Clarifications on the Design and Interpretation of Conclusions from Health Canada’s Study on Wind Turbine Noise and Health

Clarifications on the Design and Interpretation of Conclusions from Health Canada’s Study on... It has been extensively communicated that Health Canada’s Community Noise and Health Study (CNHS) did not find positive associations between wind turbine noise (WTN) levels and any of the evaluated health outcomes, beyond an increase in the prevalence of high annoyance toward several wind turbine features. The authors emphasize that this general conclusion remains bound by the study strengths and limitations. Following the publication of the CNHS findings, there has been interest among some individuals to present alternative interpretations of the results originally reported by Michaud et al. (J Acoust Soc Am 139(3):1443–1454, 2016. https://doi.org/10.1121/1.4942391). While recognizing the importance of independent scientific re-evaluation and/or reinterpretation, this commentary serves to clarify and, where necessary, correct some of the information put forward by others. One factor that has been re-evaluated by external stakeholders is the subsample of participants that comprise the lowest WTN category. In their reanalysis, they have eliminated this category, or introduced alternative comparative data. This paper identifies substantial issues associated with the re-evaluation put forth. To thoroughly address these issues and to avoid further confusion or misinterpretation, the authors of the CNHS provide a comparison between the CNHS health condition prevalence data and nationally representative health-based surveys conducted in Canada during the same calendar year. In addition, this paper responds to comments received to date on the CNHS, including the study’s age range, the generalization of findings, the provision of raw data, and conclusions on the association between WTN level and health. Keywords Canada · Community Noise and Health Study · Wind turbine · Noise · Health effects · Cross-sectional study 1 Background From 2012–2014, Health Canada, in collaboration with David S. Michaud B Statistics Canada and other external experts conducted a david.michaud@canada.ca cross-sectional study to investigate the relationship between exposure to sound levels produced from wind turbines and Environmental and Radiation Health Sciences Directorate, the extent of health effects reported by, and objectively mea- Consumer and Clinical Radiation Protection Bureau, Health Canada, 775 Brookfield Road, Ottawa, ON K1A 1C1, Canada sured in individuals living near wind turbines. In March 2016, the study findings from Health Canada’s Community Noise Population Studies Division, Biostatistics Section, Health Canada, 101 Tunney’s Pasture Driveway, Tunney’s Pasture, and Health Study (CNHS) were published in the Journal of Ottawa, ON, Canada the Acoustical Society of America as a special section on Air Health Science Division, Health Canada, 269 Laurier wind turbine noise (WTN) [1–6]. These papers followed the Avenue West, Ottawa, ON, Canada Environmental Health Program, Regulatory Operations and Department of Pediatrics, Division of Neurology, Hospital for Regions Branch, Health Canada, 1505 Barrington Street, Sick Children, University of Toronto, 555 University Avenue, Halifax, NS, Canada Toronto, ON, Canada Department of Medicine, Division of Neurology, Sunnybrook Department of Health Sciences, Carleton University, Ottawa, Health Sciences Center, University of Toronto, Toronto, ON, Canada ON, Canada 123 100 Acoustics Australia (2018) 46:99–110 CNHS’s publications related to quality of life [7] and sleep 2 Study Strengths and Limitations [8]. The study design was subjected to a rigorous peer The strengths of the CNHS are as follows: (1) large ran- review process, which included a 60-day public consulta- domly selected sample of participants (n = 1238), (2) high tion, a Research Ethics Board review, a review by Health response rate (78.9%) that did not vary by proximity to Canada’s Science Advisory Board, a review by external wind turbines in either province, (3) broadly scoped ques- experts selected by the World Health Organization (WHO), tionnaire, (4) inclusion of objectively measured endpoints of and the publication of the final planned study design [9]. stress, blood pressure, heart rate and sleep, (5) calculated Publications also reflect the assessment conducted as part of WTN levels validated with representative field measures, the independent journal review process. Discussions during and (6) an exposure–response analysis that encompassed the study design phase included the selection of a “con- a greater than 21 dB range of exposure to WTN. Despite trol group” (i.e., individuals who would have no meaningful these strengths, no single cross-sectional study should be exposure to WTN). In any epidemiological study, a con- viewed as conclusive, and all studies have some limitations. trol group is always a challenge to establish because it As part of the external peer review of the current publication, is exceedingly difficult to ensure that the only distinction it has been noted that the objectively measured outcomes between the control and exposed group is the exposure of included are not without their own shortcomings. This com- interest; in this case, WTN. In the CNHS, it was deter- ment has merit insofar as no single observation should be mined before the study was conducted that an exposure– interpreted in isolation, whether self-reported or objectively response design would be implemented. Inherent to the measured. Although each objective measure can be criti- exposure–response design is that participants are primar- cized in isolation as imperfect, there is added assurance of ily distinguished by the magnitude of their exposure to their validity insofar as they were found to be consistently WTN. Random sampling across different WTN categories related to their corresponding self-reported measures. Self- strengthens the validity of the exposure–response insofar reported high blood pressure was related to higher measured as it minimizes the likelihood that participant differences blood pressure, higher perceived stress scores were related to will bias the response to WTN at any given exposure level. higher hair cortisol concentrations and a lack of an associa- With this study design participants in the lowest WTN expo- tion between WTN levels and reported sleep disturbance (for sure group (i.e., < 25 dBA) can be viewed as a control any reason) was consistent with sleep actigraphy findings. or comparison group, even though a true control group is Cross-sectional studies are a useful and powerful epidemi- more readily established under structured laboratory condi- ological tool used to evaluate issues related to public health. tions. However, they are observational studies that collect data at a The entire sample was drawn from areas in Prince Edward specific point in time and as such they are typically limited for Island (PEI) and Ontario with similar topography, trees, hills, making causal inferences. Furthermore, they rarely have the bodies of water, climate and socioeconomic characteristics. statistical power to characterize associations between expo- Exposure to WTN levels ranged from < 25 to 46 dBA and the sures under study and health conditions that may have very distance between dwellings and turbines was between 0.25 low prevalence rates. Conclusions from the CNHS do not and 11.22 km. The study design included over-sampling in necessarily extrapolate beyond the study sample because the areas where WTN levels were highest to increase the statisti- communities in the study may have important differences cal power for detecting potential WTN-associated effects on when compared to others in Canada, or elsewhere. Similarly, sleep quality. This over-sampling was also intended to bet- the findings are representative only of areas where long-term ter characterize the exposure–response relationship between outdoor WTN levels do not exceed 46 dBA (or 63 dBC) WTN levels and various self-reported and objectively mea- [4,5] and for individuals between the ages of 18 and 79 sured outcomes in areas where potential health impacts were years. It should be acknowledged that long-term WTN calcu- more likely to be observed. As shown in Michaud et al. [1], lations do not investigate specific noise characteristics, such reproduced as Table 1, participants in the lowest WTN expo- as amplitude modulation and/or the presence of tones and are sure category had similar demographics compared to par- insensitive to very brief changes in WTN levels. Despite the ticipants in other WTN categories. Demographically, some fact that participants in the study were randomly selected, the minor differences were found with respect to age, employ- locations were not and for this reason the level of confidence ment, type of dwelling and dwelling ownership; however, for generalizing the results to other areas can only be based on with the possible exception of employment, these factors a scientific judgment regarding the level of exposure, terrain, showed no obvious pattern with WTN levels and none were climate, meteorology, and the similarity between the current strong enough to exert an influence on the overall results. study sample and others. These study limitations have been The primary distinction across the study sample, based on identified previously [10]. Thus, similar to epidemiological the data collected, was the participants’ exposure to WTN. studies conducted in other areas, this study should be viewed 123 Acoustics Australia (2018) 46:99–110 101 Table 1 Study sample characteristics reported in Health Canada’s Community Noise and Health Study Variable WTN (dBA) < 25 [25–30) [30–35) [35–40) [40–46] Overall CMH p value b b b b b b n 84 95 304 521 234 1238 Range of closest turbine (km) 2.32–11.22 1.29–4.47 0.73–2.69 0.44–1.56 0.25–1.05 Range of BNTS (dBA) 35–51 35–51 35–56 35–57 35–61 BNTS (dBA) mean (SD) 43.88 (3.43) 44.68 (2.91) 45.21 (3.60) 43.29 (4.11) 41.43 (4.21) ON 44.98 (2.88) 44.86 (2.78) 45.54 (3.31) 44.06 (3.86) 42.70 (4.25) < 0.0001 PEI 41.13 (3.18) 43.00 (3.67) 43.81 (4.38) 38.44 (1.59) 38.05 (1.00) < 0.0001 Sex n (% male) 37 (44.0) 48 (50.5) 150 (49.3) 251 (48.2) 120 (51.3) 606 (49.0) 0.4554 Age mean (SE) 49.75 (1.78) 56.38 (1.37) 52.25 (0.93) 51.26 (0.68) 50.28 (1.03) 51.61 (0.44) 0.0243 Marital status n (%) 0.2844 Married/common-law 54 (64.3) 69 (73.4) 199 (65.7) 367 (70.6) 159 (67.9) 848 (68.7) Widowed/separated/divorced 16 (19.0) 18 (19.1) 61 (20.1) 85 (16.3) 35 (15.0) 215 (17.4) Single, never been married 14 (16.7) 7 (7.4) 43 (14.2) 68 (13.1) 40 (17.1) 172 (13.9) Employed n (%) 43 (51.8) 47 (49.5) 161 (53.0) 323 (62.0) 148 (63.2) 722 (58.4) 0.0012 Level of education n (%) 0.7221 ≤ High school 45 (53.6) 52 (54.7) 167 (55.1) 280 (53.7) 134 (57.3) 678 (54.8) Trade/certificate/college 34 (40.5) 37 (38.9) 110 (36.3) 203 (39.0) 85 (36.3) 469 (37.9) University 5 (6.0) 6 (6.3) 26 (8.6) 38 (7.3) 15 (6.4) 90 (7.3) Income (x$1000) n (%) 0.8031 < 60 39 (51.3) 40 (54.8) 138 (52.5) 214 (49.1) 100 (49.3) 531 (50.5) 60–100 18 (23.7) 17 (23.3) 72 (27.4) 134 (30.7) 59 (29.1) 300 (28.5) ≥ 100 19 (25.0) 16 (21.9) 53 (20.2) 88 (20.2) 44 (21.7) 220 (20.9) Detached dwelling n (%) 59 (70.2) 84 (88.4) 267 (87.8) 506 (97.1) 216 (92.3) 1132 (91.4) e f ON 46 (76.7) 77 (89.5) 228 (93.1) 437 (97.1) 154 (90.6) 942 (93.2) < 0.0001 e f PEI 13 (54.2) 7 (77.8) 39 (66.1) 69 (97.2) 62 (96.9) 190 (83.7) < 0.0001 Property ownership n (%) 60 (71.4) 85 (89.5) 250 (82.2) 466 (89.4) 215 (91.9) 1076 (86.9) ON 45 (75.0) 78 (90.7) 215 (87.8) 399 (88.7) 157 (92.4) 894 (88.4) 0.0085 PEI 15 (62.5) 7 (77.8) 35 (59.3) 67 (94.4) 58 (90.6) 182 (80.2) < 0.0001 Façade type n (%) 0.0137 Fully bricked 20 (23.8) 30 (31.6) 85 (28.0) 138 (26.5) 67 (28.6) 340 (27.5) Partially bricked 24 (28.6) 29 (30.5) 62 (20.4) 88 (16.9) 15 (6.4) 218 (17.6) No brick/other 40 (47.6) 36 (37.9) 157 (51.6) 295 (56.6) 152 (65.0) 680 (54.9) Originally presented as Table III in reference [1] BNTS Background nighttime sound level; dBA A-weighted decibel; km kilometer; ON Ontario, PEI Prince Edward Island; SD standard deviation; SE standard error; WTN wind turbine noise The Cochran–Mantel–Haenszel (CMH) chi-square test is used to adjust for province unless otherwise indicated, p values < 0.05 are considered to be statistically significant Totals may differ due to missing data Analysis of variance (ANOVA) model Non-parametric two-way ANOVA model adjusted for province Non-detached dwellings included semi/duplex/apartment Chi-square test of independence with its numerous strengths and limitations in mind, in con- that beyond an increase in the prevalence of long-term high text of other similarly well conducted studies as well as what annoyance toward several wind turbine features [1], there was is known with respect to biologically plausible mechanisms. no evidence to support an association between WTN levels up Bearing in mind the stated strengths and limitations of to 46 dBA and any of the other self-reported or objectively the CNHS, the CNHS data support the general conclusion measured health outcomes. Reported and measured health 123 102 Acoustics Australia (2018) 46:99–110 outcomes included, but were not limited to, migraines, dizzi- 3.2 Data from the CNHS Have Not Been Provided to ness, tinnitus, blood pressure, heart disease, stress, quality the Public of life and multiple measures of sleep [1,3,7,8]. Conclusions based on objectively measured outcomes for measures of In support of transparency and scientific integrity, data orig- stress, blood pressure/heart rate and sleep have additional inating from the study are available to Canadians, other credibility insofar as they are not influenced by participant jurisdictions and interested parties through a number of awareness bias, which is always something that researchers sources that include the Statistics Canada Research Data Cen- need to consider when relying solely on self-reported mea- tres [12], and by request through the Health Canada Wind sures of health. Turbine Noise webpage (additional information) [13]. All publications are freely available as open access in scien- tific journals, and as plain language summaries on Health Canada’s Web site [13]. All data that would be required 3 Clarifications in Response to CNHS to reproduce the CNHS findings are available through the Criticisms and Misinterpretations means identified above. Data that contain information that could either be used to reveal the identity of a study partici- Constructive criticism of scientific research is encouraged pant or considered to be confidential business information is because it often stimulates improvements in future studies. not provided, consistent with requirements/exclusions under Some of the points of criticism put forward to challenge the Canada’s Statistics Act and Privacy Act. Acoustical field conclusions of the CNHS relate to issues already documented recordings made to support WTN calculations are not pro- by Health Canada as part of the acknowledged study strengths vided to the public as they contain personal conversations and limitations (see above). Other misinterpretations of the which due to the length of the recordings (over 4000 hours) CNHS findings have resulted from selective reanalysis of cannot be redacted. some of the self-reported health data by external stake- holders. The issues discussed below have been noted either 3.3 The CNHS Did Not Adequately Investigate through discussions between individuals and the CNHS prin- People Who Have Abandoned Their Homes Due cipal investigator (DSM) at scientific conferences and/or in to Health Effects Suffered Following the feedback submitted directly to Health Canada. The CNHS Installation of Wind Turbines authors’ response to each of these criticisms (summarized in bold) is presented below. This is one of the more common assertions by external stakeholders, which is, at least in part, due to the impre- 3.1 The CNHS is Flawed Because of Age Exclusions cise terminology originally used to describe addresses that were not valid dwellings and therefore considered out-of- A primary objective of the study was to assess the potential scope. The number of addresses considered out-of-scope for impacts that WTN had on measured sleep. For this reason, the sample was consistent with numbers predicted by Statis- the study design aimed to maximize the number of partici- tics Canada for a rural environment in Canada. Of the 434 pants that fell within the age range studied most frequently by out-of-scope addresses, 132 of these were identified as unoc- other researchers in this area, and in other community noise cupied for unknown reasons and were found to be randomly and sleep studies published to date. This approach would distributed across all distances studied in both provinces. be expected to increase the statistical power of the CNHS Health Canada has no way of knowing the reasons for such to detect changes in sleep, should they exist. Sleep patterns vacancies. As specified in Michaud et al. [1] locations coded among children and the elderly are sufficiently different from as out-of-scope were originally [14] assigned the follow- the study sample age group that their inclusion may have ing categories: Demolished for unknown reasons, vacant for diluted the ability to detect subtle impacts on sleep from WTN unknown reasons, unoccupied, seasonal, > 79 years of age, exposure [11]. Furthermore, the questionnaire in the study and other. In an effort to address feedback and provide further included questions that would not be suitable for minors. clarification, the categories used to define these addresses Participants above the age of 79 years were also excluded, have been more precisely defined in [1]. Specifically, loca- in part, because age-related hearing loss may influence their tions that were originally defined broadly as “unoccupied perception of WTN, and they are more likely to have other for unknown reasons” are now more precisely defined as comorbid conditions that impact sleep. Ultimately, the study 1) inhabitable dwelling not occupied at time of survey, 2) sample was limited to the age categories investigated by other newly constructed dwelling, but not yet inhabited, or 3) researchers in this area in order to maximize the possibility unoccupied trailer in vacant trailer park. Furthermore, it was of identifying impacts on sleep and other health outcomes, confirmed that 6 addresses originally identified as unoccu- should they exist. pied were in fact GPS coordinates listed in error [1]. There 123 Acoustics Australia (2018) 46:99–110 103 was no evidence in the study to support the suggestion that the appear to suggest that approximately 2 in 5 Canadians highly unoccupied dwellings have been abandoned by homeowners annoyed by road traffic noise perceive their annoyance to suffering adverse health effects from WTN exposure. have a rather strong impact on their health. However, the same survey also demonstrated that annoyance magnitude 3.4 Despite an Increase in Annoyance, the CNHS was not correlated with self-reported health status, that is, ConcludedNoAdverse Health Effects many who reported to be highly annoyed by road traffic noise also reported to be in good health [17]. Thus there No evidence was found that would reject the null hypothe- are inconsistent findings between long-term noise annoyance and potential impacts on health. Considering the compara- sis; in essence, there was no association between exposure to WTN and the self-reported or objectively measured health tively low magnitude of the aforementioned disability weight endpoints examined. However, the study did demonstrate a while noting the observations that high noise annoyance has relationship between increasing levels of WTN and annoy- been reported to be associated with other health conditions ance toward several features (including noise, perceived [16,18,19] support an interpretation of high noise annoyance indoor vibration during operations, visual impacts, shadow as a potential, but not a necessary or distinct indicator of flicker, and the aircraft warning lights on top of the tur- adverse health. Collectively, these observations may support bines) associated with wind turbines. The WHO Community decisions by jurisdictions to consider changes in the preva- Noise Guidelines list annoyance as one of the adverse health lence of community annoyance when evaluating wind turbine installation projects. effects of community noise exposure and include guidelines for annoyance that vary in level based on location and time of day [15]. In their estimation of the burden of disease 3.5 The Prevalence of Health Effects in the Lowest from environmental noise exposure, the WHO regional office WTN Category Were Inflated for Europe has assigned a “conservative” disability weight of 0.02 to long-term high (transportation) noise annoyance, Following publication of CNHS findings, there has been where 0 is equivalent to ideal health and 1 is equivalent interest among some individuals not involved in the origi- to death [16]. Although a statistical association was found nal CNHS, to reassess a sub-selection of the reported health between high WTN annoyance and several reported and mea- effects. The CNHS authors recognize the importance of sured health endpoints in the CNHS, these were unrelated to independent scientific re-evaluation and/or reinterpretation the level of WTN exposure, and there is no way of deter- however, emphasize caution when reinterpreting results that mining if these conditions may have either pre-dated, and/or have been derived through selective removal of data and were possibly exacerbated by, exposure to wind turbines [1– statistically questionable methodologies. One such reanal- 3,7]. The extent to which long-term high noise annoyance ysis involved the removal of participants from areas where may impact one’s health is uncertain. To illustrate, a national WTN levels were below 25 dBA based on a concern that the Canadian survey on road traffic noise annoyance where 2565 prevalence rates for certain health outcomes (i.e., tinnitus, respondents rated their level of annoyance toward road traf- migraines, dizziness and relative health status compared to fic noise over the previous year is highlighted. In the latter last year) were inflated and non-representative [20,21]. study, respondents assessed on an 11-point numerical scale, An alternative comparison group was comprised for one where 0 was equivalent to “no effect” and 10 was equivalent such reanalysis that included multiple data sources from the to “very strong effect,” the extent to which their annoyance USA in addition to sources from a study conducted in a toward road traffic noise was perceived to have a negative city within the province of Ontario (n=671). The Ontario impact on their health. Among respondents who rated their data were collected in 2001 and 2003 with the purpose of annoyance toward road traffic noise as high, 39% perceived assessing how self-reported health changed over time when the impact of their annoyance on their health to be equivalent the same individuals were evaluated in both surveys [22]. to 7 and above. On the other hand, only 6% of respondents Collectively, these multiple data sources have been mistak- who reported lower magnitudes of annoyance (i.e., moderate enly interpreted and presented to reflect “General Population or lower), perceived the impact on their health as 7 and above. Prevalence” data. The scientific rationale for removing the These observations imply a greater importance of “high” noise annoyance in comparison with lower magnitudes and USA data sources included Migraine Research Foundation, which reports 12% of the population suffers from migraine; however, this Vibrations/rattles during wind turbine operations were not directly statistic appears to include children, who were excluded from the CNHS. measured or modeled in the CNHS. Michaud et al [1] reported that 4.7% The same Web site indicated that 1 in 4 (or 25%) of U.S. households of participants perceived vibrations/rattles during operations, and 1.5% included an individual with migraine. Other cited USA sources include reported to be highly annoyed by vibrations/rattles. Both the perception Dizziness-and-balance.com, and Hearing Health Foundation, where the of and annoyance toward vibration/rattle were found to be statistically latter source reports that 10% of the USA adult population experienced relatedtoWTN level. tinnitus over the last 3 months. 123 104 Acoustics Australia (2018) 46:99–110 prevalence data observed in the lowest WTN exposure cat- the CNHS. These comparisons may be of interest to persons egory and then re-evaluating the recompiled data is tenuous reviewing the prevalence data published as part of the CNHS given, in part, that they were derived at different time periods [1]. for different years (almost a decade earlier) and/or nations. The Canadian Community Health Survey (CCHS) and Furthermore, the selective reanalysis of only tinnitus, dizzi- the Canadian Health Measures Survey (CHMS) [23,24] ness, migraines and relative health status compared to 1 are two large-scale population-based surveys routinely con- year earlier is inconsistent with assertions that WTN expo- ducted by Statistics Canada to collect nationally representa- sure adversely impacts a wide range of outcomes including, tive health data on Canadians. These studies are weighted but not limited to sleep, stress and anxiety, cardiovascular to account for the distribution of Canadians by sex and responses and quality of life; all of which were among the age. These surveys do not claim to be representative of 20 health conditions evaluated in the CNHS, reproduced in any particular sub-community. Individual communities may Table 2. have important differences in the sample characteristics (e.g., Several factors can reduce scientific validity when mak- health status, socioeconomic variables), which can influence ing comparisons with historical data from different stud- the reported prevalence rates. Response rates for the CCHS ies. There may be little scientific support for comparisons and CHMS tend to be lower than that observed in the CNHS between self-reported data that are collected in different (i.e., 78.9%) and therefore caution should be exercised in study populations especially when the collection periods are comparing these larger surveys with the CNHS, which is separated by several years. A more serious deterrent to such more appropriately referred to as a community study and comparisons arises where there are important differences not a national survey. Table 3 provides comparisons between between study methodologies (e.g., data collection, ques- these larger studies and the CNHS on self-reported mea- tionnaire content), which can lead to erroneous comparisons, sures of health. Potentially important differences were noted even when the endpoints assessed are similar. For example, between questionnaire content (Table 4), which should be there is a clear distinction between a question that evaluates factored into the interpretation of study differences. To our the current status of migraines or tinnitus and one that seeks knowledge, the prevalence of dizziness has not been assessed to determine if these conditions were ever experienced in in any nationally representative Canadian survey. Reported one’s lifetime. Similarly, the prevalence of a self-reported prevalence rates vary considerably depending on the type health condition is not equivalent to the prevalence of con- of dizziness evaluated, participant sex and age [25]. Indeed, sulting with a health-care professional for the same condition. several health effects are known to increase in prevalence Table 3 illustrates this difference for migraines, dizziness and with age. Since the average age in the CNHS was higher tinnitus, as reported in the CNHS. Studies like the CNHS, than the CCHS and CHMS, differences in overall prevalence that investigate the potential association between an envi- rates could potentially reflect age differences. For this reason, ronmental exposure and health, are especially sensitive to results are stratified by age category in Table 3. the possibility that publicity regarding health impacts may Finally, it should be underscored that the comparison of influence participant response (i.e., awareness bias). Strate- prevalence rates across exposure categories within any given gies to mitigate this bias in the CNHS included masking the study should consider the sample size for each exposure cat- study objective during recruitment, random sampling, a high egory. The Cochran–Mantel–Haenszel (CMH) test used in response rate and supplementing self-report with objective Michaud et al. [1] is a test used in the analysis of strat- measures. Nevertheless, awareness bias can never be fully ified categorical data. It allows an investigator to test the eliminated and is another factor to consider when comparing association between a categorical predictor or treatment and study findings that may be distorted by this bias to vary- a binary outcome such as case or control status while tak- ing degrees. No attempt was made to ensure the CNHS was ing into account the stratification of the study [26]. The test representative of a larger population as doing so is not nec- accounts for the variability or variance associated with each essary to ensure a reliable cross-sectional study. Therefore, one must avoid potential “apples to oranges” comparisons The Canadian Community Health Survey (CCHS) is a cross-sectional survey conducted by Statistics Canada to gather health-related data at as the sample population in the CNHS is not generaliz- the sub-provincial levels of geography. The CCHS relies on a large able. This has been identified by the CNHS authors as one sample (65,000) to provide reliable health-related data every 2 years. of the limitations (i.e., caution on extrapolation beyond the The CCHS produces an annual microdata file and a file combining two study sample because the communities in the study may have years of data [23]. important differences when compared to others in Canada, The Canadian Health Measures Survey (CHMS) is a survey con- ducted by Statistics Canada with the objective of collecting information or elsewhere). With these considerations in mind, this paper on Canadians’ health. The CHMS includes an in-home interview and a presents an opportunity to make some careful comparisons collection of physical measures on a wide range of outcomes, including between the CNHS and larger population-based studies that blood pressure, height, weight, bone density, hearing, and vision. The were conducted in Canada during the same calendar year as sample size of each cycle of the CHMS is approximately 5700 [24]. 123 Acoustics Australia (2018) 46:99–110 105 Table 2 Distribution of health conditions reported in Health Canada’s Community Noise and Health Study Variable n (%) WTN (dBA) < 25 [25–30) [30–35) [35–40) [40–46] Overall CMH p value b b b b b b n 84 95 304 521 234 1238 Health worse versus last year 17 (20.2) 12 (12.6) 46 (15.1) 90 (17.3) 51 (21.8) 216 (17.5) 0.1724 Migraines 18 (21.4) 24 (25.3) 56 (18.4) 134 (25.8) 57(24.4) 289 (23.4) 0.2308 Dizziness 19 (22.6) 16 (16.8) 65 (21.4) 114 (21.9) 59 (25.2) 273 (22.1) 0.2575 Tinnitus 21 (25.0) 18 (18.9) 71 (23.4) 129 (24.8) 54 (23.2) 293 (23.7) 0.7352 Chronic pain 20 (23.8) 23 (24.2) 75 (24.8) 118 (22.6) 57 (24.5) 293 (23.7) 0.8999 Asthma 8 (9.5) 12 (12.6) 22 (7.2) 43 (8.3) 16 (6.8) 101 (8.2) 0.2436 Arthritis 23 (27.4) 38 (40.0) 98 (32.2) 175 (33.7) 68 (29.1) 402 (32.5) 0.6397 High blood pressure (BP) 24 (28.6) 36 (37.9) 81 (26.8) 166 (32.0) 65 (27.8) 372 (30.2) 0.7385 Medication for high BP 26 (31.3) 34 (35.8) 84 (27.6) 163 (31.3) 63 (27.0) 370 (29.9) 0.4250 Family history of high BP 44 (52.4) 49 (53.8) 132 (45.5) 254 (50.6) 121 (53.8) 600 (50.3) 0.6015 Chronic bronchitis/emphysema/COPD 3 (3.6) 10 (10.8) 17 (5.6) 27 (5.2) 14 (6.0) 71 (5.7) 0.7676 Diabetes 7 (8.3) 8 (8.4) 33 (10.9) 46 (8.8) 19 (8.2) 113 (9.1) 0.6890 Heart disease 8 (9.5) 7 (7.4) 31 (10.2) 32 (6.1) 17 (7.3) 95 (7.7) 0.2110 Highly sleep disturbed 13 (15.7) 11 (11.6) 41 (13.5) 75 (14.5) 24 (10.3) 164 (13.3) 0.4300 Diagnosed sleep disorder 13 (15.5) 10 (10.5) 27 (8.9) 44 (8.4) 25 (10.7) 119 (9.6) 0.3102 Sleep medication 16 (19.0) 18 (18.9) 39 (12.8) 46 (8.8) 29 (12.4) 148 (12.0) 0.0083 Restless leg syndrome 7 (8.3) 16 (16.8) 37 (12.2) 81 (15.5) 33 (14.1) 174 (14.1) Restless leg syndrome (ON) 4 (6.7) 15 (17.4) 27 (11.0) 78 (17.3) 28 (16.5) 152 (15.0) 0.0629 Restless leg syndrome (PEI) 3 (12.5) 1 (11.1) 10 (16.9) 3 (4.2) 5 (7.8) 22 (9.7) 0.1628 Medication anxiety or depression 11 (13.1) 14 (14.7) 35 (11.5) 59 (11.3) 23 (9.8) 142 (11.5) 0.2470 QoL past month Poor 9 (10.8) 3 (3.2) 21 (6.9) 29 (5.6) 20 (8.6) 82 (6.6) 0.9814 Good 74 (89.2) 92 (96.8) 283 (93.1) 492 (94.4) 213 (91.4) 1154 (93.4) Satisfaction with health Dissatisfied 13 (15.5) 13 (13.7) 49 (16.1) 66 (12.7) 36 (15.4) 177 (14.3) 0.7262 Satisfied 71 (84.5) 82 (86.3) 255 (83.9) 455 (87.3) 198 (84.6) 1061 (85.7) Originally presented as Table V in reference [1] dBA A-weighted decibel; COPD chronic obstructive pulmonary disease; ON Ontario, PEI Prince Edward Island, WTN wind turbine noise The Cochran–Mantel–Haenszel (CMH) chi-square test is used to adjust for provinces unless otherwise indicated, p values < 0.05 are considered to be statistically significant Columns may not add to total due to missing data Worse consists of the two ratings: "Somewhat worse now"and "Much worse now" High sleep disturbance consists of the two ratings: “very”and “extremely” sleep disturbed Chi-square test of independence Quality of Life (QoL) and Satisfaction with Health were assessed with the two stand-alone questions on the WHOQOL-BREF. Reporting “poor” overall QoL reflects a response of “poor”or“very poor”, and “good” reflects a response of “neither poor nor good”, “good”or“very good”. Reporting “dissatisfied” overall Satisfaction with Health reflects a response of “very dissatisfied”or“dissatisfied”, and “satisfied” reflects a response of “neither satisfied nor dissatisfied”, “satisfied”or“very satisfied”. A detailed presentation of the results related to QoL is presented by reference [7] data point due to sample size within each WTN level cate- sample sizes and the error is compounded by the elimina- gory. Claims of a detectable trend in the data based solely on tion of participants from the lowest WTN exposure category. a linear regression line drawn through 4 data points are not Furthermore, a simple regression line does not adjust for supported as they do not reflect the variability (or precision) any confounding factors, an important consideration from associated with each of these data points. This variability an epidemiological point of view. For the reasons mentioned is related to the sample size in each of the WTN expo- above, the CNHS authors agree with the State of Wisconsin’s sure categories and is a necessary statistical consideration conclusion [27] that the analysis of the CNHS presented as when interpreting the CNHS data. Scientifically, the linear part of the expert testimony in [20] was inappropriate and regression model used in [20] does not take into account misleading. 123 106 Acoustics Australia (2018) 46:99–110 Table 3 Age-adjusted health conditions reported in the Canadian Health Measures Survey, Canadian Community Health Survey and Health Canada’s Community Noise and Health Study a a a CHMS (2012–2013) CCHS (2013) CNHS (2013) 18–39 40–59 60–79 18–39 40–59 60–79 18–39 40–59 60–79 n = 1258 n = 1082 n = 1049 n = 15,746 n = 17,093 n = 18,809 n = 302 n = 496 n = 440 % (95% CI) % (95% CI) % (95% CI) % (95% CI) %(95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) + ∗+ Health worse versus 12.4 (8.8, 17.2) 15.6 (13.1, 18.5) 14.7 (11.2, 19.0) 8.4 (7.8, 9.1) 11.3 (10.4, 12.3) 14.6 (13.7, 15.4) 11.6 (8.5, 15.7) 17.4 (14.3, 21.0) 21.6 (18.0, 25.7) b,c last year b E E E Asthma 13.5 (7.7, 22.5) 6.0 (3.9, 9.1) 7.9 (4.3, 14.1) 8.6 (7.9, 9.3) 7.2 (6.6, 7.9) 7.5 (6.8, 8.2) 10.3 (7.3, 14.2) 7.3 (5.3, 9.9) 7.7 (5.6, 10.6) b + ∗+ ∗+ Arthritis F 17.4 (13.2, 22.7) 38.1 (32.4, 44.2) 3.0 (2.6, 3.4) 15.8 (14.9, 16.8) 35.1 (34.0, 36.2) 7.3 (4.9, 10.8) 31.3 (27.4, 35.5) 51.1 (46.5, 55.8) b + ∗+ Hypertension F 20.8 (16.8, 25.5) 46.0 (41.4, 50.6) 2.8 (2.4, 3.2) 17.7 (16.6, 18.8) 43.0 (41.7, 44.3) 4.0 (2.3, 6.9) 24.6 (21.1, 28.6) 54.1 (49.4, 58.7) ∗+ ∗+ Medication F 17.1 (13.5, 21.3) 46.4 (41.2, 51.6) 1.5 (1.3, 1.8) 15.4 (14.4, 16.5) 44.0 (42.7, 45.3) 2.6 (1.3, 5.1) 22.6 (19.1, 26.5) 57.1 (52.4, 61.6) for hypertension Familial 48.4 (40.1, 56.8) 60.8 (53.8,67.4) 53.5 (49.3, 57.6) 45.2 (39.5, 50.9) 48.7 (44.3, 53.2) 55.7 (50.9, 60.3) hypertension E E E + ∗+ Chronic bronchitis/F2.3 (1.3, 4.0) 6.0 (3.6, 10.0) 0.2 (0.1, 0.3) 3.0 (2.6, 3.5) 6.6 (6.0, 7.2) 2.3 (1.1, 4.7) 4.0 (2.6, 6.2) 10.0 (7.5, 13.2) emphysema/ b,d COPD b E Diabetes F6.3 (3.6, 10.6) 15.5 (12.0, 19.7) 1.2 (1.0, 1.5) 6.6 (5.9, 7.3) 16.2 (15.3, 17.1) 1.7 (0.7, 3.8) 7.1 (5.1, 9.7) 16.6 (13.4, 20.4) Heart F F 11.2 (8.4, 14.8) 0.8 (0.6, 1.0) 3.1 (2.6, 3.6) 12.5 (11.8, 13.3) 1.0 (0.3, 2.9) 5.6 (3.9, 8.0) 14.6 (11.6, 18.2) disease b,e,f Self-rated quality of life E E E Poor 3.9 (2.0, 7.6) 10.4 (7.1, 15.1) 5.4 (3.6, 8.1) 2.3 (1.1, 4.7) 7.9 (5.8, 10.6) 8.2 (6.0, 11.1) Good 96.1 (92.4, 98.0) 89.6 (84.9, 92.9) 94.6 (91.9, 96.4) 97.7 (95.3, 98.9) 92.1 (89.4, 94.2) 91.8 (88.9, 94.0) b,f,g Health in general Poor/dissatisfied 8.6 (6.1, 12.0) 12.4 (9.1, 16.7) 16.1 (12.8, 20.1) 5.7 (5.1, 6.2) 12.0 (11.1, 13.0) 18.0 (17.1, 19.0) 8.6 (5.9, 12.3) 16.7 (13.7, 20.3) 15.5 (12.4, 19.1) Good/satisfied 91.4 (88.0, 93.9) 87.6 (83.3, 90.9) 83.9 (79.9, 87.2) 94.3 (93.8, 94.9) 88.0 (87.0, 88.9) 82.0 (81.0, 82.9) 91.4 (87.7, 94.1) 83.3 (79.7, 86.3) 84.5 (80.9, 87.6) Migraine 30.6 (25.6, 36.0) 25.2 (21.6, 29.2) 16.4 (13.2, 20.1) headaches + + Consulted 11.7 (10.8, 12.5) 12.4 (11.5, 13.4) 6.3 (5.7, 6.9) 17.6 (13.7, 22.3) 12.1 (9.5, 15.3) 9.3 (6.9, 12.4) for migraine headaches Acoustics Australia (2018) 46:99–110 107 Table 3 continued a a a CHMS (2012–2013) CCHS (2013) CNHS (2013) 18–39 40–59 60–79 18–39 40–59 60–79 18–39 40–59 60–79 n = 1258 n = 1082 n = 1049 n = 15,746 n = 17,093 n = 18,809 n = 302 n = 496 n = 440 % (95% CI) % (95% CI) % (95% CI) % (95% CI) %(95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) b ∗ ∗ Tinnitus 37.8 (31.4, 44.7) 32.5 (25.2, 40.7) 33.9 (29.4, 38.7) 17.9 (14.0, 22.6) 25.0 (21.4, 29.0) 26.2 (22.3, 30.5) Consulted 5.3 (3.3, 8.4) 8.1 (6.0, 10.8) 13.7 (10.8, 17.2) for tinnitus Dizziness 21.2 (17.0, 26.1) 22.8 (19.3, 26.7) 21.8 (18.2, 25.9) Consulted 11.6 (8.5, 15.7) 14.3 (11.5, 17.7) 14.5 (11.6, 18.1) for dizziness CI confidence interval; CNHS Community Noise and Health Study; CCHS Canadian Community Health Survey; CHMS Canadian Health Measures Survey, COPD Chronic obstructive pulmonary disease High sampling variability; use with caution Suppressed due to extreme sampling variability Mean age is 45.4, 45.6 and 51.6 years old for CHMS, CCHS and CNHS, respectively Health endpoint unrelated to WTN exposure in the CNHS Worse consists of the two ratings: “Somewhat worse now” and “Much worse now” Reflects a "yes" to any of the three conditions in CHMS "Poor" includes categories "poor" and "fair"; "Good" includes categories "good", "very good" and "excellent" "Poor" includes categories "very poor" and "poor"; "Good" includes categories "neither poor nor good", "good" and "very good" Reporting “dissatisfied” overall Satisfaction with Health reflects a response of “very dissatisfied” or “dissatisfied”, and “satisfied” reflects a response of “neither satisfied nor dissatisfied”, “satisfied” or “very satisfied" significantly different from CHMS at the 5% level significantly different from CCHS at the 5% level Note, important differences in the questions used to evaluate each health condition in the three studies are shown in Table 4 108 Acoustics Australia (2018) 46:99–110 Table 4 Disparity between questions used to assess the same health conditions in the Canadian Health Measures Survey, Canadian Community Health Survey and Health Canada’s Community Noise and Health Study Health condition CNHS question CHMS question CCHS question Health worse versus last year Compared to one year ago, how would Same question as CNHS Same question as CNHS you say your health is now? Is it…? Much better now, Somewhat better now, About the same, Somewhat worse now, Much worse now Asthma Do you have asthma? Same question as CNHS Same question as CNHS Arthritis Do you have arthritis? Same question as CNHS Do you have arthritis, excluding fibromyalgia? Hypertension Do you have high blood pressure? Same question as CNHS Same question as CNHS Medication for hypertension In the past month, have you taken any Same question as CNHS Same question as CNHS medicine for high blood pressure? Familial hypertension (risk Is there a history of high blood pressure Has anyone in your immedi- Not evaluated factor) in your family? ate family ever had high blood pressure, excluding during preg- nancy? Chronic bronchitis/ Do you have chronic bronchitis, Do you have chronic bronchitis? Same question as CNHS ac emphysema/COPD emphysema or chronic obstructive pul- Do you have emphysema? monary disease? Do you have chronic obstructive pulmonary disease? Diabetes Do you have diabetes? Same question as CNHS Same question as CNHS Heart disease Do you have heart disease? Same question as CNHS Same question as CNHS Self-rated quality of life In the past month, how would you rate Would you rate your quality of Not evaluated your quality of life? Very poor, Poor, life as…? Poor, Fair, Good, Very Neither poor nor good, Good, Very good, Excellent good Health in general In the past month, how satisfied were In general, would you say your In general, would you say you with your health? Very dissatisfied, health is…? Poor, Fair, Good, your health is…? Poor, Fair, Dissatisfied, Neither satisfied nor dis- Very good, Excellent Good, Very good, Excellent satisfied, Satisfied, Very Satisfied d,e Migraine headaches In the last 12 months, have you experi- Not evaluated Do you have migraine enced frequent migraines or headaches headaches? (includes nausea, vomiting, sensitivity to light and sound)? d,f Tinnitus In the last 12 months, have you expe- Now I’d like to ask you about Not evaluated rienced ringing, buzzing or whistling tinnitus. Tinnitus is the pres- sounds in your ears for no reason? ence of hissing, buzzing, ring- ing, rushing or roaring sounds in your ears when there is no other sound around you. Have you ever experienced tinnitus? Dizziness In the last 12 months, have you experi- Not evaluated Not evaluated enced dizziness? CNHS Community Noise and Health Study; CCHS Canadian Community Health Survey; CHMS Canadian Health Measures Survey, COPD Chronic obstructive pulmonary disease In CNHS, CHMS and CCHS these questions are preceeded by : "We are interested in "long-term conditions" which are expected to last or have already lasted 6 months or more and that have been diagnosed by a health professional" In CCHS, this question is skipped if the answer to high blood pressure is "no" Question asked only if age ≥ 35 in CCHS, and questions on emphysema and COPD if age ≥ 30 in CHMS In CNHS, if participants reported the condition, a follow-up question asked if they consulted with a health-care professional regarding the condition. Results originally reported by Michaud et al. [1] were for self-reported prevalence as only a subsample would be expected to consult with a health-care professional about the condition In CCHS, response to question on migraines was preceeded with the following reminder by the interviewer: "Remember, we’re interested in conditions diagnosed by a health professional and that are expected to last or have already lasted 6 months or more" Tinnitus was evaluated in CHMS with response categories that permitted comparison to CNHS, i.e., experienced within the last year 123 Acoustics Australia (2018) 46:99–110 109 4 Concluding Remarks to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. No single study, regardless how comprehensive, can be expected to provide all of the answers to the many questions References that exist in any given area of research and any study should be considered in the context of the broader evidence base. 1. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., Knowledge gained through science is incremental, advanced Than, J., Guay, M., Denning, A., McGuire, D., Bower, T., Lavigne, through replication and consistency in observed outcomes E., Murray, B.J., Weiss, S.K., van den Berg, F.: Exposure to wind from studies that employ different study designs and meth- turbine noise: Perceptual responses and reported health effects. J. Acoust. Soc. Am. 139(3), 1443–1454 (2016). https://doi.org/10. ods of exposure assessment. The CNHS results support an 1121/1.4942391 association between increasing WTN levels and an increase 2. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., in the prevalence of annoyance toward various wind turbine Than, J., Guay, M., Bower, T., Denning, A., Lavigne, E., Whelan, features. As noted in the discussion of limitations, cross- C., Janssen, S.A., van den Berg, F.: Personal and situational vari- ables associated with wind turbine noise annoyance. J. Acoust. sectional studies are not sufficient to establish causality, yet Soc. Am. 139(3), 1455–1466 (2016). https://doi.org/10.1121/1. they do have the strength of assessing multiple outcomes and exposures at the same time in large populations over 3. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., short periods of time. For this reason, they often serve as the Than, J., Guay, M., Denning, A., Bower, T., Villeneuve, P., Russell, E., Koren, G., van den Berg, F.: Self-reported and measured stress basis for hypothesis testing in follow-up case–control and related responses associated with exposure to wind turbine noise. cohort studies. The correlations that were observed between J. Acoust. Soc. Am. 139(3), 1467–1497 (2016). https://doi.org/10. reported high WTN annoyance and some of the self-reported 1121/1.4942402 and measured health outcomes are not sufficient, in isola- 4. Keith, S.E., Feder, K., Voicescu, S., Soukhovtsev, V., Denning, A., Tsang, J., Broner, N., Richarz, W., van den Berg, F.: Wind turbine tion, to suggest that high degrees of WTN annoyance cause sound power measurements. J. Acoust. Soc. Am. 139(3), 1431– these outcomes (or vice versa). These associations may be 1435 (2016). https://doi.org/10.1121/1.4942405 influenced by other risk factors that are unaccounted for in a 5. Keith, S.E., Feder, K., Voicescu, S., Soukhovtsev, V., Denning, A., single cross-sectional study, or by design biases (e.g., uses of Tsang, J., Broner, N., Richarz, W., van den Berg, F.: Wind turbine sound pressure level calculations at dwellings. J. Acoust. Soc. Am. self-reported data, participation bias). Should an association 139(3), 1436–1442 (2016). https://doi.org/10.1121/1.4942404 between high WTN annoyance and adverse health outcomes 6. Voicescu, S., Michaud, D.S., Feder, K., Marro, L., Than, J., Guay, be established in the future, efforts to minimize annoyance be M., Denning, A., Bower, T., van den Berg, F., Broner, N., Lavigne, it from acoustical, or non-acoustical features (e.g., blinking E.: Estimating annoyance to calculated wind turbine shadow flicker is improved when variables associated with wind turbine noise light, shadow flicker mitigation) may be supported on those exposure are considered. J. Acoust. Soc. Am. 139(3), 1480–1492 grounds. (2016). https://doi.org/10.1121/1.4942403 The motivation behind the current commentary was to 7. Feder, K., Michaud, D.S., Keith, S.E., Voicescu, S.A., Marro, L., provide a formal response to feedback that has been received Than, J., Guay, M., Denning, A., Bower, T.J., Lavigne, E., Whelan, C., van den Berg, F.: An assessment of quality of life using the now that more than a year has passed since the primary WHOQOL-BREF among participants living in the vicinity of wind research findings from the CNHS have been published. It turbines. Environ. Res. 142, 227–238 (2015). https://doi.org/10. also serves to identify several issues of concern around the 1016/j.envres.2015.06.043 reanalysis of the CNHS data presented as part of the expert 8. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., Than, J., Guay, M., Denning, A., Murray, B.J., Weiss, S.K., Vil- testimony in [20]. Publishing in a special issue dedicated to leneuve, P.J., van den Berg, F., Bower, T.: Effects of wind turbine WTN should broaden the reach of this response within the noise on self-reported and objective measures of sleep. SLEEP scientific community, highlighting important epidemiolog- 39(1), 97–109 (2016). https://doi.org/10.5665/sleep.5326 ical principles that need to be considered when evaluating 9. Michaud, D.S., Keith, S.E., Feder, K., Soukhovtsev, V., Marro, L., Denning, A., McGuire, D., Broner, N., Richarz, W., Tsang, health studies. Our intention is that this commentary may J., Legault, S., Poulin, D., Bryan, S., Duddeck, C., Lavigne, E., serve as an acknowledgement of, and a collective response Villeneuve, P.J., Leroux, T., Weiss, S.K., Murray, B.J., Bower, T.: to, a range of issues that will undoubtedly remain relevant Self-reported and objectively measured health indicators among a so long as the CNHS continues to inform discussions that sample of Canadians living within the vicinity of industrial wind turbines: Social survey and sound level modeling methodology. surround the growing science base related to WTN exposure Noise News Int. 21, 14–27 (2013) and human health. 10. Health Canada: Health Impacts and Exposure to Sound From Wind Turbines: Updated Research Design and Sound Exposure Assessment. https://www.canada.ca/en/health-canada/services/ Open Access This article is distributed under the terms of the Creative environmental-workplace-health/consultations/health-impacts- Commons Attribution 4.0 International License (http://creativecomm exposure-sound-wind-turbines-updated-research-design-sound- ons.org/licenses/by/4.0/), which permits unrestricted use, distribution, exposure-assessment.html (2014). Accessed 14 Nov 2017 and reproduction in any medium, provided you give appropriate credit 11. Ohayon, M.M., Carskadon, M.A., Guilleminault, C., Vitiello, M.W.: Meta-analysis of quantitative sleep parameters from child- 123 110 Acoustics Australia (2018) 46:99–110 hood to old age in healthy individuals: developing normative 21. Nissenbaum, M.A.: Industrial wind turbines and adverse health sleep values across the human lifespan. SLEEP 27(7), 1255–1273 effects: where we are, where we need to go, and the need for reg- (2004). https://doi.org/10.1093/sleep/27.7.1255 ulations and predictive models to recognize human physiology. 12. Statistics Canada Research Data Centre reference. The Research J. Acoust. Soc. Am. 139, 2149 (2016). https://doi.org/10.1121/1. Data Centres (RDC) Program. http://www.statcan.gc.ca/eng/rdc/ 4950348 index (2017). Accessed 14 Nov 2017 22. Wilson, K., Elliott, S.J., Eyles, J.D., Keller-Olaman, S.J.: Factors 13. Health Canada: Wind turbine noise. https://www.canada.ca/en/ affecting change over time in self-reported health. Can. J. Pub- health-canada/services/environmental-workplace-health/noise/ lic Health 98(2) 154–158 (2007). https://www.ncbi.nlm.nih.gov/ wind-turbine-noise.html (2014). Accessed 14 Nov 2017 pubmed/17441542 14. Health Canada: Wind turbine Noise and Health Study: Summary 23. Statistics Canada: Canadian Community Health Survey - of Results. https://www.canada.ca/en/health-canada/services/ Annual Component (CCHS). https://www.statcan.gc.ca/eng/ environmental-workplace-health/noise/wind-turbine-noise/ survey/household/3226 (2015). Accessed 15 Jan 2018 wind-turbine-noise-health-study-summary-results.html (2014). 24. Statistics Canada: Ongoing. Canadian Health Measures Survey. Accessed 14 Nov 2017 http://www.statcan.gc.ca/eng/survey/household/5071 (Ongoing). 15. World Health Organization: Guidelines for Community Noise. Accessed 14 Nov 2017 WHO, Geneva (1999) 25. Bittar, R.S.M., Oiticia, J., Bottino, M.A., Ganança, F.F., Dimitrov, 16. World Health Organization (WHO): Burden of disease from envi- R.: Population epidemiological study on the prevalence of dizziness ronmental noise. Quantification of healthy life years lost in Europe. in the city of São Paulo. Braz. J. Otorhinolaryngol. 79(6), 688–698 Fritschi, L., Brown, A.L., Kim, R., Schwela, D., Kephalopolous, (2013). https://doi.org/10.5935/1808-8694.20130127 S. (eds.). Bonn: World Health Organization, Regional Office for 26. Agresti, A.: Categorical Data Analysis (PDF), pp. 231–232. Wiley, Europe (2011) Hooken (2002). ISBN 0-471-36093-7 17. Michaud, D.S., Keith, S.E., McMurchy, D.: Annoyance and dis- 27. Public Service Commission of Wisconsin (PSCW): Application turbance of daily activities from road traffic noise in Canada. of Highland Wind Farm, LLC, for a certificate of public con- J. Acoust. Soc. Am. 123(2), 784–792 (2008). https://doi.org/10. venience and necessity to construct a 102.5 MW wind electric 1121/1.2821984 generation facility and associated electric facilities, to be located 18. Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, in the towns of Forest and Cylon, St. Croix County, Wisconsin. S., Stansfeld, S.: Auditory and non-auditory effects of noise on Final Decision on Remand. http://apps.psc.wi.gov/vs2015/ERF_ health. Lancet 383(9925), 1325–1332 (2014). https://doi.org/10. view/viewdoc.aspx?docid=290039 (2016). Accessed 14 Nov 2017 1016/S0140-6736(13)61613-X 19. Niemann, H., Bonnefoy, X., Braubach, M., Hecht, K., Maschke, C., Rodrigues, C., Röbbel, N.: Noise-induced annoyance and morbidity results from the pan-European LARES study. Noise Health 8(31), 63–79 (2006). http://www.noiseandhealth.org/text. asp?2006/8/31/63/33537 20. Public Service Commission of Wisconsin (PSCW): Appli- cation of Highland Wind Farm, LLC, for a certificate of public convenience and necessity to construct a 102.5 MW wind electric generation facility and associated electric facil- ities, to be located in the towns of Forest and Cylon, St. Croix County, Wisconsin. Expert statement of Richard James. Docket No.: 2525-CE-100. http://apps.psc.wi.gov/vs2015/ERF_ view/viewdoc.aspx?docid=284895. Accessed 14 Nov 2017 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acoustics Australia Springer Journals

Clarifications on the Design and Interpretation of Conclusions from Health Canada’s Study on Wind Turbine Noise and Health

Free
12 pages
Loading next page...
 
/lp/springer_journal/clarifications-on-the-design-and-interpretation-of-conclusions-from-fABno1osow
Publisher
Springer Singapore
Copyright
Copyright © 2018 by The Author(s)
Subject
Engineering; Engineering Acoustics; Acoustics; Noise Control
ISSN
0814-6039
eISSN
1839-2571
D.O.I.
10.1007/s40857-017-0125-4
Publisher site
See Article on Publisher Site

Abstract

It has been extensively communicated that Health Canada’s Community Noise and Health Study (CNHS) did not find positive associations between wind turbine noise (WTN) levels and any of the evaluated health outcomes, beyond an increase in the prevalence of high annoyance toward several wind turbine features. The authors emphasize that this general conclusion remains bound by the study strengths and limitations. Following the publication of the CNHS findings, there has been interest among some individuals to present alternative interpretations of the results originally reported by Michaud et al. (J Acoust Soc Am 139(3):1443–1454, 2016. https://doi.org/10.1121/1.4942391). While recognizing the importance of independent scientific re-evaluation and/or reinterpretation, this commentary serves to clarify and, where necessary, correct some of the information put forward by others. One factor that has been re-evaluated by external stakeholders is the subsample of participants that comprise the lowest WTN category. In their reanalysis, they have eliminated this category, or introduced alternative comparative data. This paper identifies substantial issues associated with the re-evaluation put forth. To thoroughly address these issues and to avoid further confusion or misinterpretation, the authors of the CNHS provide a comparison between the CNHS health condition prevalence data and nationally representative health-based surveys conducted in Canada during the same calendar year. In addition, this paper responds to comments received to date on the CNHS, including the study’s age range, the generalization of findings, the provision of raw data, and conclusions on the association between WTN level and health. Keywords Canada · Community Noise and Health Study · Wind turbine · Noise · Health effects · Cross-sectional study 1 Background From 2012–2014, Health Canada, in collaboration with David S. Michaud B Statistics Canada and other external experts conducted a david.michaud@canada.ca cross-sectional study to investigate the relationship between exposure to sound levels produced from wind turbines and Environmental and Radiation Health Sciences Directorate, the extent of health effects reported by, and objectively mea- Consumer and Clinical Radiation Protection Bureau, Health Canada, 775 Brookfield Road, Ottawa, ON K1A 1C1, Canada sured in individuals living near wind turbines. In March 2016, the study findings from Health Canada’s Community Noise Population Studies Division, Biostatistics Section, Health Canada, 101 Tunney’s Pasture Driveway, Tunney’s Pasture, and Health Study (CNHS) were published in the Journal of Ottawa, ON, Canada the Acoustical Society of America as a special section on Air Health Science Division, Health Canada, 269 Laurier wind turbine noise (WTN) [1–6]. These papers followed the Avenue West, Ottawa, ON, Canada Environmental Health Program, Regulatory Operations and Department of Pediatrics, Division of Neurology, Hospital for Regions Branch, Health Canada, 1505 Barrington Street, Sick Children, University of Toronto, 555 University Avenue, Halifax, NS, Canada Toronto, ON, Canada Department of Medicine, Division of Neurology, Sunnybrook Department of Health Sciences, Carleton University, Ottawa, Health Sciences Center, University of Toronto, Toronto, ON, Canada ON, Canada 123 100 Acoustics Australia (2018) 46:99–110 CNHS’s publications related to quality of life [7] and sleep 2 Study Strengths and Limitations [8]. The study design was subjected to a rigorous peer The strengths of the CNHS are as follows: (1) large ran- review process, which included a 60-day public consulta- domly selected sample of participants (n = 1238), (2) high tion, a Research Ethics Board review, a review by Health response rate (78.9%) that did not vary by proximity to Canada’s Science Advisory Board, a review by external wind turbines in either province, (3) broadly scoped ques- experts selected by the World Health Organization (WHO), tionnaire, (4) inclusion of objectively measured endpoints of and the publication of the final planned study design [9]. stress, blood pressure, heart rate and sleep, (5) calculated Publications also reflect the assessment conducted as part of WTN levels validated with representative field measures, the independent journal review process. Discussions during and (6) an exposure–response analysis that encompassed the study design phase included the selection of a “con- a greater than 21 dB range of exposure to WTN. Despite trol group” (i.e., individuals who would have no meaningful these strengths, no single cross-sectional study should be exposure to WTN). In any epidemiological study, a con- viewed as conclusive, and all studies have some limitations. trol group is always a challenge to establish because it As part of the external peer review of the current publication, is exceedingly difficult to ensure that the only distinction it has been noted that the objectively measured outcomes between the control and exposed group is the exposure of included are not without their own shortcomings. This com- interest; in this case, WTN. In the CNHS, it was deter- ment has merit insofar as no single observation should be mined before the study was conducted that an exposure– interpreted in isolation, whether self-reported or objectively response design would be implemented. Inherent to the measured. Although each objective measure can be criti- exposure–response design is that participants are primar- cized in isolation as imperfect, there is added assurance of ily distinguished by the magnitude of their exposure to their validity insofar as they were found to be consistently WTN. Random sampling across different WTN categories related to their corresponding self-reported measures. Self- strengthens the validity of the exposure–response insofar reported high blood pressure was related to higher measured as it minimizes the likelihood that participant differences blood pressure, higher perceived stress scores were related to will bias the response to WTN at any given exposure level. higher hair cortisol concentrations and a lack of an associa- With this study design participants in the lowest WTN expo- tion between WTN levels and reported sleep disturbance (for sure group (i.e., < 25 dBA) can be viewed as a control any reason) was consistent with sleep actigraphy findings. or comparison group, even though a true control group is Cross-sectional studies are a useful and powerful epidemi- more readily established under structured laboratory condi- ological tool used to evaluate issues related to public health. tions. However, they are observational studies that collect data at a The entire sample was drawn from areas in Prince Edward specific point in time and as such they are typically limited for Island (PEI) and Ontario with similar topography, trees, hills, making causal inferences. Furthermore, they rarely have the bodies of water, climate and socioeconomic characteristics. statistical power to characterize associations between expo- Exposure to WTN levels ranged from < 25 to 46 dBA and the sures under study and health conditions that may have very distance between dwellings and turbines was between 0.25 low prevalence rates. Conclusions from the CNHS do not and 11.22 km. The study design included over-sampling in necessarily extrapolate beyond the study sample because the areas where WTN levels were highest to increase the statisti- communities in the study may have important differences cal power for detecting potential WTN-associated effects on when compared to others in Canada, or elsewhere. Similarly, sleep quality. This over-sampling was also intended to bet- the findings are representative only of areas where long-term ter characterize the exposure–response relationship between outdoor WTN levels do not exceed 46 dBA (or 63 dBC) WTN levels and various self-reported and objectively mea- [4,5] and for individuals between the ages of 18 and 79 sured outcomes in areas where potential health impacts were years. It should be acknowledged that long-term WTN calcu- more likely to be observed. As shown in Michaud et al. [1], lations do not investigate specific noise characteristics, such reproduced as Table 1, participants in the lowest WTN expo- as amplitude modulation and/or the presence of tones and are sure category had similar demographics compared to par- insensitive to very brief changes in WTN levels. Despite the ticipants in other WTN categories. Demographically, some fact that participants in the study were randomly selected, the minor differences were found with respect to age, employ- locations were not and for this reason the level of confidence ment, type of dwelling and dwelling ownership; however, for generalizing the results to other areas can only be based on with the possible exception of employment, these factors a scientific judgment regarding the level of exposure, terrain, showed no obvious pattern with WTN levels and none were climate, meteorology, and the similarity between the current strong enough to exert an influence on the overall results. study sample and others. These study limitations have been The primary distinction across the study sample, based on identified previously [10]. Thus, similar to epidemiological the data collected, was the participants’ exposure to WTN. studies conducted in other areas, this study should be viewed 123 Acoustics Australia (2018) 46:99–110 101 Table 1 Study sample characteristics reported in Health Canada’s Community Noise and Health Study Variable WTN (dBA) < 25 [25–30) [30–35) [35–40) [40–46] Overall CMH p value b b b b b b n 84 95 304 521 234 1238 Range of closest turbine (km) 2.32–11.22 1.29–4.47 0.73–2.69 0.44–1.56 0.25–1.05 Range of BNTS (dBA) 35–51 35–51 35–56 35–57 35–61 BNTS (dBA) mean (SD) 43.88 (3.43) 44.68 (2.91) 45.21 (3.60) 43.29 (4.11) 41.43 (4.21) ON 44.98 (2.88) 44.86 (2.78) 45.54 (3.31) 44.06 (3.86) 42.70 (4.25) < 0.0001 PEI 41.13 (3.18) 43.00 (3.67) 43.81 (4.38) 38.44 (1.59) 38.05 (1.00) < 0.0001 Sex n (% male) 37 (44.0) 48 (50.5) 150 (49.3) 251 (48.2) 120 (51.3) 606 (49.0) 0.4554 Age mean (SE) 49.75 (1.78) 56.38 (1.37) 52.25 (0.93) 51.26 (0.68) 50.28 (1.03) 51.61 (0.44) 0.0243 Marital status n (%) 0.2844 Married/common-law 54 (64.3) 69 (73.4) 199 (65.7) 367 (70.6) 159 (67.9) 848 (68.7) Widowed/separated/divorced 16 (19.0) 18 (19.1) 61 (20.1) 85 (16.3) 35 (15.0) 215 (17.4) Single, never been married 14 (16.7) 7 (7.4) 43 (14.2) 68 (13.1) 40 (17.1) 172 (13.9) Employed n (%) 43 (51.8) 47 (49.5) 161 (53.0) 323 (62.0) 148 (63.2) 722 (58.4) 0.0012 Level of education n (%) 0.7221 ≤ High school 45 (53.6) 52 (54.7) 167 (55.1) 280 (53.7) 134 (57.3) 678 (54.8) Trade/certificate/college 34 (40.5) 37 (38.9) 110 (36.3) 203 (39.0) 85 (36.3) 469 (37.9) University 5 (6.0) 6 (6.3) 26 (8.6) 38 (7.3) 15 (6.4) 90 (7.3) Income (x$1000) n (%) 0.8031 < 60 39 (51.3) 40 (54.8) 138 (52.5) 214 (49.1) 100 (49.3) 531 (50.5) 60–100 18 (23.7) 17 (23.3) 72 (27.4) 134 (30.7) 59 (29.1) 300 (28.5) ≥ 100 19 (25.0) 16 (21.9) 53 (20.2) 88 (20.2) 44 (21.7) 220 (20.9) Detached dwelling n (%) 59 (70.2) 84 (88.4) 267 (87.8) 506 (97.1) 216 (92.3) 1132 (91.4) e f ON 46 (76.7) 77 (89.5) 228 (93.1) 437 (97.1) 154 (90.6) 942 (93.2) < 0.0001 e f PEI 13 (54.2) 7 (77.8) 39 (66.1) 69 (97.2) 62 (96.9) 190 (83.7) < 0.0001 Property ownership n (%) 60 (71.4) 85 (89.5) 250 (82.2) 466 (89.4) 215 (91.9) 1076 (86.9) ON 45 (75.0) 78 (90.7) 215 (87.8) 399 (88.7) 157 (92.4) 894 (88.4) 0.0085 PEI 15 (62.5) 7 (77.8) 35 (59.3) 67 (94.4) 58 (90.6) 182 (80.2) < 0.0001 Façade type n (%) 0.0137 Fully bricked 20 (23.8) 30 (31.6) 85 (28.0) 138 (26.5) 67 (28.6) 340 (27.5) Partially bricked 24 (28.6) 29 (30.5) 62 (20.4) 88 (16.9) 15 (6.4) 218 (17.6) No brick/other 40 (47.6) 36 (37.9) 157 (51.6) 295 (56.6) 152 (65.0) 680 (54.9) Originally presented as Table III in reference [1] BNTS Background nighttime sound level; dBA A-weighted decibel; km kilometer; ON Ontario, PEI Prince Edward Island; SD standard deviation; SE standard error; WTN wind turbine noise The Cochran–Mantel–Haenszel (CMH) chi-square test is used to adjust for province unless otherwise indicated, p values < 0.05 are considered to be statistically significant Totals may differ due to missing data Analysis of variance (ANOVA) model Non-parametric two-way ANOVA model adjusted for province Non-detached dwellings included semi/duplex/apartment Chi-square test of independence with its numerous strengths and limitations in mind, in con- that beyond an increase in the prevalence of long-term high text of other similarly well conducted studies as well as what annoyance toward several wind turbine features [1], there was is known with respect to biologically plausible mechanisms. no evidence to support an association between WTN levels up Bearing in mind the stated strengths and limitations of to 46 dBA and any of the other self-reported or objectively the CNHS, the CNHS data support the general conclusion measured health outcomes. Reported and measured health 123 102 Acoustics Australia (2018) 46:99–110 outcomes included, but were not limited to, migraines, dizzi- 3.2 Data from the CNHS Have Not Been Provided to ness, tinnitus, blood pressure, heart disease, stress, quality the Public of life and multiple measures of sleep [1,3,7,8]. Conclusions based on objectively measured outcomes for measures of In support of transparency and scientific integrity, data orig- stress, blood pressure/heart rate and sleep have additional inating from the study are available to Canadians, other credibility insofar as they are not influenced by participant jurisdictions and interested parties through a number of awareness bias, which is always something that researchers sources that include the Statistics Canada Research Data Cen- need to consider when relying solely on self-reported mea- tres [12], and by request through the Health Canada Wind sures of health. Turbine Noise webpage (additional information) [13]. All publications are freely available as open access in scien- tific journals, and as plain language summaries on Health Canada’s Web site [13]. All data that would be required 3 Clarifications in Response to CNHS to reproduce the CNHS findings are available through the Criticisms and Misinterpretations means identified above. Data that contain information that could either be used to reveal the identity of a study partici- Constructive criticism of scientific research is encouraged pant or considered to be confidential business information is because it often stimulates improvements in future studies. not provided, consistent with requirements/exclusions under Some of the points of criticism put forward to challenge the Canada’s Statistics Act and Privacy Act. Acoustical field conclusions of the CNHS relate to issues already documented recordings made to support WTN calculations are not pro- by Health Canada as part of the acknowledged study strengths vided to the public as they contain personal conversations and limitations (see above). Other misinterpretations of the which due to the length of the recordings (over 4000 hours) CNHS findings have resulted from selective reanalysis of cannot be redacted. some of the self-reported health data by external stake- holders. The issues discussed below have been noted either 3.3 The CNHS Did Not Adequately Investigate through discussions between individuals and the CNHS prin- People Who Have Abandoned Their Homes Due cipal investigator (DSM) at scientific conferences and/or in to Health Effects Suffered Following the feedback submitted directly to Health Canada. The CNHS Installation of Wind Turbines authors’ response to each of these criticisms (summarized in bold) is presented below. This is one of the more common assertions by external stakeholders, which is, at least in part, due to the impre- 3.1 The CNHS is Flawed Because of Age Exclusions cise terminology originally used to describe addresses that were not valid dwellings and therefore considered out-of- A primary objective of the study was to assess the potential scope. The number of addresses considered out-of-scope for impacts that WTN had on measured sleep. For this reason, the sample was consistent with numbers predicted by Statis- the study design aimed to maximize the number of partici- tics Canada for a rural environment in Canada. Of the 434 pants that fell within the age range studied most frequently by out-of-scope addresses, 132 of these were identified as unoc- other researchers in this area, and in other community noise cupied for unknown reasons and were found to be randomly and sleep studies published to date. This approach would distributed across all distances studied in both provinces. be expected to increase the statistical power of the CNHS Health Canada has no way of knowing the reasons for such to detect changes in sleep, should they exist. Sleep patterns vacancies. As specified in Michaud et al. [1] locations coded among children and the elderly are sufficiently different from as out-of-scope were originally [14] assigned the follow- the study sample age group that their inclusion may have ing categories: Demolished for unknown reasons, vacant for diluted the ability to detect subtle impacts on sleep from WTN unknown reasons, unoccupied, seasonal, > 79 years of age, exposure [11]. Furthermore, the questionnaire in the study and other. In an effort to address feedback and provide further included questions that would not be suitable for minors. clarification, the categories used to define these addresses Participants above the age of 79 years were also excluded, have been more precisely defined in [1]. Specifically, loca- in part, because age-related hearing loss may influence their tions that were originally defined broadly as “unoccupied perception of WTN, and they are more likely to have other for unknown reasons” are now more precisely defined as comorbid conditions that impact sleep. Ultimately, the study 1) inhabitable dwelling not occupied at time of survey, 2) sample was limited to the age categories investigated by other newly constructed dwelling, but not yet inhabited, or 3) researchers in this area in order to maximize the possibility unoccupied trailer in vacant trailer park. Furthermore, it was of identifying impacts on sleep and other health outcomes, confirmed that 6 addresses originally identified as unoccu- should they exist. pied were in fact GPS coordinates listed in error [1]. There 123 Acoustics Australia (2018) 46:99–110 103 was no evidence in the study to support the suggestion that the appear to suggest that approximately 2 in 5 Canadians highly unoccupied dwellings have been abandoned by homeowners annoyed by road traffic noise perceive their annoyance to suffering adverse health effects from WTN exposure. have a rather strong impact on their health. However, the same survey also demonstrated that annoyance magnitude 3.4 Despite an Increase in Annoyance, the CNHS was not correlated with self-reported health status, that is, ConcludedNoAdverse Health Effects many who reported to be highly annoyed by road traffic noise also reported to be in good health [17]. Thus there No evidence was found that would reject the null hypothe- are inconsistent findings between long-term noise annoyance and potential impacts on health. Considering the compara- sis; in essence, there was no association between exposure to WTN and the self-reported or objectively measured health tively low magnitude of the aforementioned disability weight endpoints examined. However, the study did demonstrate a while noting the observations that high noise annoyance has relationship between increasing levels of WTN and annoy- been reported to be associated with other health conditions ance toward several features (including noise, perceived [16,18,19] support an interpretation of high noise annoyance indoor vibration during operations, visual impacts, shadow as a potential, but not a necessary or distinct indicator of flicker, and the aircraft warning lights on top of the tur- adverse health. Collectively, these observations may support bines) associated with wind turbines. The WHO Community decisions by jurisdictions to consider changes in the preva- Noise Guidelines list annoyance as one of the adverse health lence of community annoyance when evaluating wind turbine installation projects. effects of community noise exposure and include guidelines for annoyance that vary in level based on location and time of day [15]. In their estimation of the burden of disease 3.5 The Prevalence of Health Effects in the Lowest from environmental noise exposure, the WHO regional office WTN Category Were Inflated for Europe has assigned a “conservative” disability weight of 0.02 to long-term high (transportation) noise annoyance, Following publication of CNHS findings, there has been where 0 is equivalent to ideal health and 1 is equivalent interest among some individuals not involved in the origi- to death [16]. Although a statistical association was found nal CNHS, to reassess a sub-selection of the reported health between high WTN annoyance and several reported and mea- effects. The CNHS authors recognize the importance of sured health endpoints in the CNHS, these were unrelated to independent scientific re-evaluation and/or reinterpretation the level of WTN exposure, and there is no way of deter- however, emphasize caution when reinterpreting results that mining if these conditions may have either pre-dated, and/or have been derived through selective removal of data and were possibly exacerbated by, exposure to wind turbines [1– statistically questionable methodologies. One such reanal- 3,7]. The extent to which long-term high noise annoyance ysis involved the removal of participants from areas where may impact one’s health is uncertain. To illustrate, a national WTN levels were below 25 dBA based on a concern that the Canadian survey on road traffic noise annoyance where 2565 prevalence rates for certain health outcomes (i.e., tinnitus, respondents rated their level of annoyance toward road traf- migraines, dizziness and relative health status compared to fic noise over the previous year is highlighted. In the latter last year) were inflated and non-representative [20,21]. study, respondents assessed on an 11-point numerical scale, An alternative comparison group was comprised for one where 0 was equivalent to “no effect” and 10 was equivalent such reanalysis that included multiple data sources from the to “very strong effect,” the extent to which their annoyance USA in addition to sources from a study conducted in a toward road traffic noise was perceived to have a negative city within the province of Ontario (n=671). The Ontario impact on their health. Among respondents who rated their data were collected in 2001 and 2003 with the purpose of annoyance toward road traffic noise as high, 39% perceived assessing how self-reported health changed over time when the impact of their annoyance on their health to be equivalent the same individuals were evaluated in both surveys [22]. to 7 and above. On the other hand, only 6% of respondents Collectively, these multiple data sources have been mistak- who reported lower magnitudes of annoyance (i.e., moderate enly interpreted and presented to reflect “General Population or lower), perceived the impact on their health as 7 and above. Prevalence” data. The scientific rationale for removing the These observations imply a greater importance of “high” noise annoyance in comparison with lower magnitudes and USA data sources included Migraine Research Foundation, which reports 12% of the population suffers from migraine; however, this Vibrations/rattles during wind turbine operations were not directly statistic appears to include children, who were excluded from the CNHS. measured or modeled in the CNHS. Michaud et al [1] reported that 4.7% The same Web site indicated that 1 in 4 (or 25%) of U.S. households of participants perceived vibrations/rattles during operations, and 1.5% included an individual with migraine. Other cited USA sources include reported to be highly annoyed by vibrations/rattles. Both the perception Dizziness-and-balance.com, and Hearing Health Foundation, where the of and annoyance toward vibration/rattle were found to be statistically latter source reports that 10% of the USA adult population experienced relatedtoWTN level. tinnitus over the last 3 months. 123 104 Acoustics Australia (2018) 46:99–110 prevalence data observed in the lowest WTN exposure cat- the CNHS. These comparisons may be of interest to persons egory and then re-evaluating the recompiled data is tenuous reviewing the prevalence data published as part of the CNHS given, in part, that they were derived at different time periods [1]. for different years (almost a decade earlier) and/or nations. The Canadian Community Health Survey (CCHS) and Furthermore, the selective reanalysis of only tinnitus, dizzi- the Canadian Health Measures Survey (CHMS) [23,24] ness, migraines and relative health status compared to 1 are two large-scale population-based surveys routinely con- year earlier is inconsistent with assertions that WTN expo- ducted by Statistics Canada to collect nationally representa- sure adversely impacts a wide range of outcomes including, tive health data on Canadians. These studies are weighted but not limited to sleep, stress and anxiety, cardiovascular to account for the distribution of Canadians by sex and responses and quality of life; all of which were among the age. These surveys do not claim to be representative of 20 health conditions evaluated in the CNHS, reproduced in any particular sub-community. Individual communities may Table 2. have important differences in the sample characteristics (e.g., Several factors can reduce scientific validity when mak- health status, socioeconomic variables), which can influence ing comparisons with historical data from different stud- the reported prevalence rates. Response rates for the CCHS ies. There may be little scientific support for comparisons and CHMS tend to be lower than that observed in the CNHS between self-reported data that are collected in different (i.e., 78.9%) and therefore caution should be exercised in study populations especially when the collection periods are comparing these larger surveys with the CNHS, which is separated by several years. A more serious deterrent to such more appropriately referred to as a community study and comparisons arises where there are important differences not a national survey. Table 3 provides comparisons between between study methodologies (e.g., data collection, ques- these larger studies and the CNHS on self-reported mea- tionnaire content), which can lead to erroneous comparisons, sures of health. Potentially important differences were noted even when the endpoints assessed are similar. For example, between questionnaire content (Table 4), which should be there is a clear distinction between a question that evaluates factored into the interpretation of study differences. To our the current status of migraines or tinnitus and one that seeks knowledge, the prevalence of dizziness has not been assessed to determine if these conditions were ever experienced in in any nationally representative Canadian survey. Reported one’s lifetime. Similarly, the prevalence of a self-reported prevalence rates vary considerably depending on the type health condition is not equivalent to the prevalence of con- of dizziness evaluated, participant sex and age [25]. Indeed, sulting with a health-care professional for the same condition. several health effects are known to increase in prevalence Table 3 illustrates this difference for migraines, dizziness and with age. Since the average age in the CNHS was higher tinnitus, as reported in the CNHS. Studies like the CNHS, than the CCHS and CHMS, differences in overall prevalence that investigate the potential association between an envi- rates could potentially reflect age differences. For this reason, ronmental exposure and health, are especially sensitive to results are stratified by age category in Table 3. the possibility that publicity regarding health impacts may Finally, it should be underscored that the comparison of influence participant response (i.e., awareness bias). Strate- prevalence rates across exposure categories within any given gies to mitigate this bias in the CNHS included masking the study should consider the sample size for each exposure cat- study objective during recruitment, random sampling, a high egory. The Cochran–Mantel–Haenszel (CMH) test used in response rate and supplementing self-report with objective Michaud et al. [1] is a test used in the analysis of strat- measures. Nevertheless, awareness bias can never be fully ified categorical data. It allows an investigator to test the eliminated and is another factor to consider when comparing association between a categorical predictor or treatment and study findings that may be distorted by this bias to vary- a binary outcome such as case or control status while tak- ing degrees. No attempt was made to ensure the CNHS was ing into account the stratification of the study [26]. The test representative of a larger population as doing so is not nec- accounts for the variability or variance associated with each essary to ensure a reliable cross-sectional study. Therefore, one must avoid potential “apples to oranges” comparisons The Canadian Community Health Survey (CCHS) is a cross-sectional survey conducted by Statistics Canada to gather health-related data at as the sample population in the CNHS is not generaliz- the sub-provincial levels of geography. The CCHS relies on a large able. This has been identified by the CNHS authors as one sample (65,000) to provide reliable health-related data every 2 years. of the limitations (i.e., caution on extrapolation beyond the The CCHS produces an annual microdata file and a file combining two study sample because the communities in the study may have years of data [23]. important differences when compared to others in Canada, The Canadian Health Measures Survey (CHMS) is a survey con- ducted by Statistics Canada with the objective of collecting information or elsewhere). With these considerations in mind, this paper on Canadians’ health. The CHMS includes an in-home interview and a presents an opportunity to make some careful comparisons collection of physical measures on a wide range of outcomes, including between the CNHS and larger population-based studies that blood pressure, height, weight, bone density, hearing, and vision. The were conducted in Canada during the same calendar year as sample size of each cycle of the CHMS is approximately 5700 [24]. 123 Acoustics Australia (2018) 46:99–110 105 Table 2 Distribution of health conditions reported in Health Canada’s Community Noise and Health Study Variable n (%) WTN (dBA) < 25 [25–30) [30–35) [35–40) [40–46] Overall CMH p value b b b b b b n 84 95 304 521 234 1238 Health worse versus last year 17 (20.2) 12 (12.6) 46 (15.1) 90 (17.3) 51 (21.8) 216 (17.5) 0.1724 Migraines 18 (21.4) 24 (25.3) 56 (18.4) 134 (25.8) 57(24.4) 289 (23.4) 0.2308 Dizziness 19 (22.6) 16 (16.8) 65 (21.4) 114 (21.9) 59 (25.2) 273 (22.1) 0.2575 Tinnitus 21 (25.0) 18 (18.9) 71 (23.4) 129 (24.8) 54 (23.2) 293 (23.7) 0.7352 Chronic pain 20 (23.8) 23 (24.2) 75 (24.8) 118 (22.6) 57 (24.5) 293 (23.7) 0.8999 Asthma 8 (9.5) 12 (12.6) 22 (7.2) 43 (8.3) 16 (6.8) 101 (8.2) 0.2436 Arthritis 23 (27.4) 38 (40.0) 98 (32.2) 175 (33.7) 68 (29.1) 402 (32.5) 0.6397 High blood pressure (BP) 24 (28.6) 36 (37.9) 81 (26.8) 166 (32.0) 65 (27.8) 372 (30.2) 0.7385 Medication for high BP 26 (31.3) 34 (35.8) 84 (27.6) 163 (31.3) 63 (27.0) 370 (29.9) 0.4250 Family history of high BP 44 (52.4) 49 (53.8) 132 (45.5) 254 (50.6) 121 (53.8) 600 (50.3) 0.6015 Chronic bronchitis/emphysema/COPD 3 (3.6) 10 (10.8) 17 (5.6) 27 (5.2) 14 (6.0) 71 (5.7) 0.7676 Diabetes 7 (8.3) 8 (8.4) 33 (10.9) 46 (8.8) 19 (8.2) 113 (9.1) 0.6890 Heart disease 8 (9.5) 7 (7.4) 31 (10.2) 32 (6.1) 17 (7.3) 95 (7.7) 0.2110 Highly sleep disturbed 13 (15.7) 11 (11.6) 41 (13.5) 75 (14.5) 24 (10.3) 164 (13.3) 0.4300 Diagnosed sleep disorder 13 (15.5) 10 (10.5) 27 (8.9) 44 (8.4) 25 (10.7) 119 (9.6) 0.3102 Sleep medication 16 (19.0) 18 (18.9) 39 (12.8) 46 (8.8) 29 (12.4) 148 (12.0) 0.0083 Restless leg syndrome 7 (8.3) 16 (16.8) 37 (12.2) 81 (15.5) 33 (14.1) 174 (14.1) Restless leg syndrome (ON) 4 (6.7) 15 (17.4) 27 (11.0) 78 (17.3) 28 (16.5) 152 (15.0) 0.0629 Restless leg syndrome (PEI) 3 (12.5) 1 (11.1) 10 (16.9) 3 (4.2) 5 (7.8) 22 (9.7) 0.1628 Medication anxiety or depression 11 (13.1) 14 (14.7) 35 (11.5) 59 (11.3) 23 (9.8) 142 (11.5) 0.2470 QoL past month Poor 9 (10.8) 3 (3.2) 21 (6.9) 29 (5.6) 20 (8.6) 82 (6.6) 0.9814 Good 74 (89.2) 92 (96.8) 283 (93.1) 492 (94.4) 213 (91.4) 1154 (93.4) Satisfaction with health Dissatisfied 13 (15.5) 13 (13.7) 49 (16.1) 66 (12.7) 36 (15.4) 177 (14.3) 0.7262 Satisfied 71 (84.5) 82 (86.3) 255 (83.9) 455 (87.3) 198 (84.6) 1061 (85.7) Originally presented as Table V in reference [1] dBA A-weighted decibel; COPD chronic obstructive pulmonary disease; ON Ontario, PEI Prince Edward Island, WTN wind turbine noise The Cochran–Mantel–Haenszel (CMH) chi-square test is used to adjust for provinces unless otherwise indicated, p values < 0.05 are considered to be statistically significant Columns may not add to total due to missing data Worse consists of the two ratings: "Somewhat worse now"and "Much worse now" High sleep disturbance consists of the two ratings: “very”and “extremely” sleep disturbed Chi-square test of independence Quality of Life (QoL) and Satisfaction with Health were assessed with the two stand-alone questions on the WHOQOL-BREF. Reporting “poor” overall QoL reflects a response of “poor”or“very poor”, and “good” reflects a response of “neither poor nor good”, “good”or“very good”. Reporting “dissatisfied” overall Satisfaction with Health reflects a response of “very dissatisfied”or“dissatisfied”, and “satisfied” reflects a response of “neither satisfied nor dissatisfied”, “satisfied”or“very satisfied”. A detailed presentation of the results related to QoL is presented by reference [7] data point due to sample size within each WTN level cate- sample sizes and the error is compounded by the elimina- gory. Claims of a detectable trend in the data based solely on tion of participants from the lowest WTN exposure category. a linear regression line drawn through 4 data points are not Furthermore, a simple regression line does not adjust for supported as they do not reflect the variability (or precision) any confounding factors, an important consideration from associated with each of these data points. This variability an epidemiological point of view. For the reasons mentioned is related to the sample size in each of the WTN expo- above, the CNHS authors agree with the State of Wisconsin’s sure categories and is a necessary statistical consideration conclusion [27] that the analysis of the CNHS presented as when interpreting the CNHS data. Scientifically, the linear part of the expert testimony in [20] was inappropriate and regression model used in [20] does not take into account misleading. 123 106 Acoustics Australia (2018) 46:99–110 Table 3 Age-adjusted health conditions reported in the Canadian Health Measures Survey, Canadian Community Health Survey and Health Canada’s Community Noise and Health Study a a a CHMS (2012–2013) CCHS (2013) CNHS (2013) 18–39 40–59 60–79 18–39 40–59 60–79 18–39 40–59 60–79 n = 1258 n = 1082 n = 1049 n = 15,746 n = 17,093 n = 18,809 n = 302 n = 496 n = 440 % (95% CI) % (95% CI) % (95% CI) % (95% CI) %(95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) + ∗+ Health worse versus 12.4 (8.8, 17.2) 15.6 (13.1, 18.5) 14.7 (11.2, 19.0) 8.4 (7.8, 9.1) 11.3 (10.4, 12.3) 14.6 (13.7, 15.4) 11.6 (8.5, 15.7) 17.4 (14.3, 21.0) 21.6 (18.0, 25.7) b,c last year b E E E Asthma 13.5 (7.7, 22.5) 6.0 (3.9, 9.1) 7.9 (4.3, 14.1) 8.6 (7.9, 9.3) 7.2 (6.6, 7.9) 7.5 (6.8, 8.2) 10.3 (7.3, 14.2) 7.3 (5.3, 9.9) 7.7 (5.6, 10.6) b + ∗+ ∗+ Arthritis F 17.4 (13.2, 22.7) 38.1 (32.4, 44.2) 3.0 (2.6, 3.4) 15.8 (14.9, 16.8) 35.1 (34.0, 36.2) 7.3 (4.9, 10.8) 31.3 (27.4, 35.5) 51.1 (46.5, 55.8) b + ∗+ Hypertension F 20.8 (16.8, 25.5) 46.0 (41.4, 50.6) 2.8 (2.4, 3.2) 17.7 (16.6, 18.8) 43.0 (41.7, 44.3) 4.0 (2.3, 6.9) 24.6 (21.1, 28.6) 54.1 (49.4, 58.7) ∗+ ∗+ Medication F 17.1 (13.5, 21.3) 46.4 (41.2, 51.6) 1.5 (1.3, 1.8) 15.4 (14.4, 16.5) 44.0 (42.7, 45.3) 2.6 (1.3, 5.1) 22.6 (19.1, 26.5) 57.1 (52.4, 61.6) for hypertension Familial 48.4 (40.1, 56.8) 60.8 (53.8,67.4) 53.5 (49.3, 57.6) 45.2 (39.5, 50.9) 48.7 (44.3, 53.2) 55.7 (50.9, 60.3) hypertension E E E + ∗+ Chronic bronchitis/F2.3 (1.3, 4.0) 6.0 (3.6, 10.0) 0.2 (0.1, 0.3) 3.0 (2.6, 3.5) 6.6 (6.0, 7.2) 2.3 (1.1, 4.7) 4.0 (2.6, 6.2) 10.0 (7.5, 13.2) emphysema/ b,d COPD b E Diabetes F6.3 (3.6, 10.6) 15.5 (12.0, 19.7) 1.2 (1.0, 1.5) 6.6 (5.9, 7.3) 16.2 (15.3, 17.1) 1.7 (0.7, 3.8) 7.1 (5.1, 9.7) 16.6 (13.4, 20.4) Heart F F 11.2 (8.4, 14.8) 0.8 (0.6, 1.0) 3.1 (2.6, 3.6) 12.5 (11.8, 13.3) 1.0 (0.3, 2.9) 5.6 (3.9, 8.0) 14.6 (11.6, 18.2) disease b,e,f Self-rated quality of life E E E Poor 3.9 (2.0, 7.6) 10.4 (7.1, 15.1) 5.4 (3.6, 8.1) 2.3 (1.1, 4.7) 7.9 (5.8, 10.6) 8.2 (6.0, 11.1) Good 96.1 (92.4, 98.0) 89.6 (84.9, 92.9) 94.6 (91.9, 96.4) 97.7 (95.3, 98.9) 92.1 (89.4, 94.2) 91.8 (88.9, 94.0) b,f,g Health in general Poor/dissatisfied 8.6 (6.1, 12.0) 12.4 (9.1, 16.7) 16.1 (12.8, 20.1) 5.7 (5.1, 6.2) 12.0 (11.1, 13.0) 18.0 (17.1, 19.0) 8.6 (5.9, 12.3) 16.7 (13.7, 20.3) 15.5 (12.4, 19.1) Good/satisfied 91.4 (88.0, 93.9) 87.6 (83.3, 90.9) 83.9 (79.9, 87.2) 94.3 (93.8, 94.9) 88.0 (87.0, 88.9) 82.0 (81.0, 82.9) 91.4 (87.7, 94.1) 83.3 (79.7, 86.3) 84.5 (80.9, 87.6) Migraine 30.6 (25.6, 36.0) 25.2 (21.6, 29.2) 16.4 (13.2, 20.1) headaches + + Consulted 11.7 (10.8, 12.5) 12.4 (11.5, 13.4) 6.3 (5.7, 6.9) 17.6 (13.7, 22.3) 12.1 (9.5, 15.3) 9.3 (6.9, 12.4) for migraine headaches Acoustics Australia (2018) 46:99–110 107 Table 3 continued a a a CHMS (2012–2013) CCHS (2013) CNHS (2013) 18–39 40–59 60–79 18–39 40–59 60–79 18–39 40–59 60–79 n = 1258 n = 1082 n = 1049 n = 15,746 n = 17,093 n = 18,809 n = 302 n = 496 n = 440 % (95% CI) % (95% CI) % (95% CI) % (95% CI) %(95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) b ∗ ∗ Tinnitus 37.8 (31.4, 44.7) 32.5 (25.2, 40.7) 33.9 (29.4, 38.7) 17.9 (14.0, 22.6) 25.0 (21.4, 29.0) 26.2 (22.3, 30.5) Consulted 5.3 (3.3, 8.4) 8.1 (6.0, 10.8) 13.7 (10.8, 17.2) for tinnitus Dizziness 21.2 (17.0, 26.1) 22.8 (19.3, 26.7) 21.8 (18.2, 25.9) Consulted 11.6 (8.5, 15.7) 14.3 (11.5, 17.7) 14.5 (11.6, 18.1) for dizziness CI confidence interval; CNHS Community Noise and Health Study; CCHS Canadian Community Health Survey; CHMS Canadian Health Measures Survey, COPD Chronic obstructive pulmonary disease High sampling variability; use with caution Suppressed due to extreme sampling variability Mean age is 45.4, 45.6 and 51.6 years old for CHMS, CCHS and CNHS, respectively Health endpoint unrelated to WTN exposure in the CNHS Worse consists of the two ratings: “Somewhat worse now” and “Much worse now” Reflects a "yes" to any of the three conditions in CHMS "Poor" includes categories "poor" and "fair"; "Good" includes categories "good", "very good" and "excellent" "Poor" includes categories "very poor" and "poor"; "Good" includes categories "neither poor nor good", "good" and "very good" Reporting “dissatisfied” overall Satisfaction with Health reflects a response of “very dissatisfied” or “dissatisfied”, and “satisfied” reflects a response of “neither satisfied nor dissatisfied”, “satisfied” or “very satisfied" significantly different from CHMS at the 5% level significantly different from CCHS at the 5% level Note, important differences in the questions used to evaluate each health condition in the three studies are shown in Table 4 108 Acoustics Australia (2018) 46:99–110 Table 4 Disparity between questions used to assess the same health conditions in the Canadian Health Measures Survey, Canadian Community Health Survey and Health Canada’s Community Noise and Health Study Health condition CNHS question CHMS question CCHS question Health worse versus last year Compared to one year ago, how would Same question as CNHS Same question as CNHS you say your health is now? Is it…? Much better now, Somewhat better now, About the same, Somewhat worse now, Much worse now Asthma Do you have asthma? Same question as CNHS Same question as CNHS Arthritis Do you have arthritis? Same question as CNHS Do you have arthritis, excluding fibromyalgia? Hypertension Do you have high blood pressure? Same question as CNHS Same question as CNHS Medication for hypertension In the past month, have you taken any Same question as CNHS Same question as CNHS medicine for high blood pressure? Familial hypertension (risk Is there a history of high blood pressure Has anyone in your immedi- Not evaluated factor) in your family? ate family ever had high blood pressure, excluding during preg- nancy? Chronic bronchitis/ Do you have chronic bronchitis, Do you have chronic bronchitis? Same question as CNHS ac emphysema/COPD emphysema or chronic obstructive pul- Do you have emphysema? monary disease? Do you have chronic obstructive pulmonary disease? Diabetes Do you have diabetes? Same question as CNHS Same question as CNHS Heart disease Do you have heart disease? Same question as CNHS Same question as CNHS Self-rated quality of life In the past month, how would you rate Would you rate your quality of Not evaluated your quality of life? Very poor, Poor, life as…? Poor, Fair, Good, Very Neither poor nor good, Good, Very good, Excellent good Health in general In the past month, how satisfied were In general, would you say your In general, would you say you with your health? Very dissatisfied, health is…? Poor, Fair, Good, your health is…? Poor, Fair, Dissatisfied, Neither satisfied nor dis- Very good, Excellent Good, Very good, Excellent satisfied, Satisfied, Very Satisfied d,e Migraine headaches In the last 12 months, have you experi- Not evaluated Do you have migraine enced frequent migraines or headaches headaches? (includes nausea, vomiting, sensitivity to light and sound)? d,f Tinnitus In the last 12 months, have you expe- Now I’d like to ask you about Not evaluated rienced ringing, buzzing or whistling tinnitus. Tinnitus is the pres- sounds in your ears for no reason? ence of hissing, buzzing, ring- ing, rushing or roaring sounds in your ears when there is no other sound around you. Have you ever experienced tinnitus? Dizziness In the last 12 months, have you experi- Not evaluated Not evaluated enced dizziness? CNHS Community Noise and Health Study; CCHS Canadian Community Health Survey; CHMS Canadian Health Measures Survey, COPD Chronic obstructive pulmonary disease In CNHS, CHMS and CCHS these questions are preceeded by : "We are interested in "long-term conditions" which are expected to last or have already lasted 6 months or more and that have been diagnosed by a health professional" In CCHS, this question is skipped if the answer to high blood pressure is "no" Question asked only if age ≥ 35 in CCHS, and questions on emphysema and COPD if age ≥ 30 in CHMS In CNHS, if participants reported the condition, a follow-up question asked if they consulted with a health-care professional regarding the condition. Results originally reported by Michaud et al. [1] were for self-reported prevalence as only a subsample would be expected to consult with a health-care professional about the condition In CCHS, response to question on migraines was preceeded with the following reminder by the interviewer: "Remember, we’re interested in conditions diagnosed by a health professional and that are expected to last or have already lasted 6 months or more" Tinnitus was evaluated in CHMS with response categories that permitted comparison to CNHS, i.e., experienced within the last year 123 Acoustics Australia (2018) 46:99–110 109 4 Concluding Remarks to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. No single study, regardless how comprehensive, can be expected to provide all of the answers to the many questions References that exist in any given area of research and any study should be considered in the context of the broader evidence base. 1. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., Knowledge gained through science is incremental, advanced Than, J., Guay, M., Denning, A., McGuire, D., Bower, T., Lavigne, through replication and consistency in observed outcomes E., Murray, B.J., Weiss, S.K., van den Berg, F.: Exposure to wind from studies that employ different study designs and meth- turbine noise: Perceptual responses and reported health effects. J. Acoust. Soc. Am. 139(3), 1443–1454 (2016). https://doi.org/10. ods of exposure assessment. The CNHS results support an 1121/1.4942391 association between increasing WTN levels and an increase 2. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., in the prevalence of annoyance toward various wind turbine Than, J., Guay, M., Bower, T., Denning, A., Lavigne, E., Whelan, features. As noted in the discussion of limitations, cross- C., Janssen, S.A., van den Berg, F.: Personal and situational vari- ables associated with wind turbine noise annoyance. J. Acoust. sectional studies are not sufficient to establish causality, yet Soc. Am. 139(3), 1455–1466 (2016). https://doi.org/10.1121/1. they do have the strength of assessing multiple outcomes and exposures at the same time in large populations over 3. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., short periods of time. For this reason, they often serve as the Than, J., Guay, M., Denning, A., Bower, T., Villeneuve, P., Russell, E., Koren, G., van den Berg, F.: Self-reported and measured stress basis for hypothesis testing in follow-up case–control and related responses associated with exposure to wind turbine noise. cohort studies. The correlations that were observed between J. Acoust. Soc. Am. 139(3), 1467–1497 (2016). https://doi.org/10. reported high WTN annoyance and some of the self-reported 1121/1.4942402 and measured health outcomes are not sufficient, in isola- 4. Keith, S.E., Feder, K., Voicescu, S., Soukhovtsev, V., Denning, A., Tsang, J., Broner, N., Richarz, W., van den Berg, F.: Wind turbine tion, to suggest that high degrees of WTN annoyance cause sound power measurements. J. Acoust. Soc. Am. 139(3), 1431– these outcomes (or vice versa). These associations may be 1435 (2016). https://doi.org/10.1121/1.4942405 influenced by other risk factors that are unaccounted for in a 5. Keith, S.E., Feder, K., Voicescu, S., Soukhovtsev, V., Denning, A., single cross-sectional study, or by design biases (e.g., uses of Tsang, J., Broner, N., Richarz, W., van den Berg, F.: Wind turbine sound pressure level calculations at dwellings. J. Acoust. Soc. Am. self-reported data, participation bias). Should an association 139(3), 1436–1442 (2016). https://doi.org/10.1121/1.4942404 between high WTN annoyance and adverse health outcomes 6. Voicescu, S., Michaud, D.S., Feder, K., Marro, L., Than, J., Guay, be established in the future, efforts to minimize annoyance be M., Denning, A., Bower, T., van den Berg, F., Broner, N., Lavigne, it from acoustical, or non-acoustical features (e.g., blinking E.: Estimating annoyance to calculated wind turbine shadow flicker is improved when variables associated with wind turbine noise light, shadow flicker mitigation) may be supported on those exposure are considered. J. Acoust. Soc. Am. 139(3), 1480–1492 grounds. (2016). https://doi.org/10.1121/1.4942403 The motivation behind the current commentary was to 7. Feder, K., Michaud, D.S., Keith, S.E., Voicescu, S.A., Marro, L., provide a formal response to feedback that has been received Than, J., Guay, M., Denning, A., Bower, T.J., Lavigne, E., Whelan, C., van den Berg, F.: An assessment of quality of life using the now that more than a year has passed since the primary WHOQOL-BREF among participants living in the vicinity of wind research findings from the CNHS have been published. It turbines. Environ. Res. 142, 227–238 (2015). https://doi.org/10. also serves to identify several issues of concern around the 1016/j.envres.2015.06.043 reanalysis of the CNHS data presented as part of the expert 8. Michaud, D.S., Feder, K., Keith, S.E., Voicescu, S.A., Marro, L., Than, J., Guay, M., Denning, A., Murray, B.J., Weiss, S.K., Vil- testimony in [20]. Publishing in a special issue dedicated to leneuve, P.J., van den Berg, F., Bower, T.: Effects of wind turbine WTN should broaden the reach of this response within the noise on self-reported and objective measures of sleep. SLEEP scientific community, highlighting important epidemiolog- 39(1), 97–109 (2016). https://doi.org/10.5665/sleep.5326 ical principles that need to be considered when evaluating 9. Michaud, D.S., Keith, S.E., Feder, K., Soukhovtsev, V., Marro, L., Denning, A., McGuire, D., Broner, N., Richarz, W., Tsang, health studies. Our intention is that this commentary may J., Legault, S., Poulin, D., Bryan, S., Duddeck, C., Lavigne, E., serve as an acknowledgement of, and a collective response Villeneuve, P.J., Leroux, T., Weiss, S.K., Murray, B.J., Bower, T.: to, a range of issues that will undoubtedly remain relevant Self-reported and objectively measured health indicators among a so long as the CNHS continues to inform discussions that sample of Canadians living within the vicinity of industrial wind turbines: Social survey and sound level modeling methodology. surround the growing science base related to WTN exposure Noise News Int. 21, 14–27 (2013) and human health. 10. Health Canada: Health Impacts and Exposure to Sound From Wind Turbines: Updated Research Design and Sound Exposure Assessment. https://www.canada.ca/en/health-canada/services/ Open Access This article is distributed under the terms of the Creative environmental-workplace-health/consultations/health-impacts- Commons Attribution 4.0 International License (http://creativecomm exposure-sound-wind-turbines-updated-research-design-sound- ons.org/licenses/by/4.0/), which permits unrestricted use, distribution, exposure-assessment.html (2014). Accessed 14 Nov 2017 and reproduction in any medium, provided you give appropriate credit 11. Ohayon, M.M., Carskadon, M.A., Guilleminault, C., Vitiello, M.W.: Meta-analysis of quantitative sleep parameters from child- 123 110 Acoustics Australia (2018) 46:99–110 hood to old age in healthy individuals: developing normative 21. Nissenbaum, M.A.: Industrial wind turbines and adverse health sleep values across the human lifespan. SLEEP 27(7), 1255–1273 effects: where we are, where we need to go, and the need for reg- (2004). https://doi.org/10.1093/sleep/27.7.1255 ulations and predictive models to recognize human physiology. 12. Statistics Canada Research Data Centre reference. The Research J. Acoust. Soc. Am. 139, 2149 (2016). https://doi.org/10.1121/1. Data Centres (RDC) Program. http://www.statcan.gc.ca/eng/rdc/ 4950348 index (2017). Accessed 14 Nov 2017 22. Wilson, K., Elliott, S.J., Eyles, J.D., Keller-Olaman, S.J.: Factors 13. Health Canada: Wind turbine noise. https://www.canada.ca/en/ affecting change over time in self-reported health. Can. J. Pub- health-canada/services/environmental-workplace-health/noise/ lic Health 98(2) 154–158 (2007). https://www.ncbi.nlm.nih.gov/ wind-turbine-noise.html (2014). Accessed 14 Nov 2017 pubmed/17441542 14. Health Canada: Wind turbine Noise and Health Study: Summary 23. Statistics Canada: Canadian Community Health Survey - of Results. https://www.canada.ca/en/health-canada/services/ Annual Component (CCHS). https://www.statcan.gc.ca/eng/ environmental-workplace-health/noise/wind-turbine-noise/ survey/household/3226 (2015). Accessed 15 Jan 2018 wind-turbine-noise-health-study-summary-results.html (2014). 24. Statistics Canada: Ongoing. Canadian Health Measures Survey. Accessed 14 Nov 2017 http://www.statcan.gc.ca/eng/survey/household/5071 (Ongoing). 15. World Health Organization: Guidelines for Community Noise. Accessed 14 Nov 2017 WHO, Geneva (1999) 25. Bittar, R.S.M., Oiticia, J., Bottino, M.A., Ganança, F.F., Dimitrov, 16. World Health Organization (WHO): Burden of disease from envi- R.: Population epidemiological study on the prevalence of dizziness ronmental noise. Quantification of healthy life years lost in Europe. in the city of São Paulo. Braz. J. Otorhinolaryngol. 79(6), 688–698 Fritschi, L., Brown, A.L., Kim, R., Schwela, D., Kephalopolous, (2013). https://doi.org/10.5935/1808-8694.20130127 S. (eds.). Bonn: World Health Organization, Regional Office for 26. Agresti, A.: Categorical Data Analysis (PDF), pp. 231–232. Wiley, Europe (2011) Hooken (2002). ISBN 0-471-36093-7 17. Michaud, D.S., Keith, S.E., McMurchy, D.: Annoyance and dis- 27. Public Service Commission of Wisconsin (PSCW): Application turbance of daily activities from road traffic noise in Canada. of Highland Wind Farm, LLC, for a certificate of public con- J. Acoust. Soc. Am. 123(2), 784–792 (2008). https://doi.org/10. venience and necessity to construct a 102.5 MW wind electric 1121/1.2821984 generation facility and associated electric facilities, to be located 18. Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, in the towns of Forest and Cylon, St. Croix County, Wisconsin. S., Stansfeld, S.: Auditory and non-auditory effects of noise on Final Decision on Remand. http://apps.psc.wi.gov/vs2015/ERF_ health. Lancet 383(9925), 1325–1332 (2014). https://doi.org/10. view/viewdoc.aspx?docid=290039 (2016). Accessed 14 Nov 2017 1016/S0140-6736(13)61613-X 19. Niemann, H., Bonnefoy, X., Braubach, M., Hecht, K., Maschke, C., Rodrigues, C., Röbbel, N.: Noise-induced annoyance and morbidity results from the pan-European LARES study. Noise Health 8(31), 63–79 (2006). http://www.noiseandhealth.org/text. asp?2006/8/31/63/33537 20. Public Service Commission of Wisconsin (PSCW): Appli- cation of Highland Wind Farm, LLC, for a certificate of public convenience and necessity to construct a 102.5 MW wind electric generation facility and associated electric facil- ities, to be located in the towns of Forest and Cylon, St. Croix County, Wisconsin. Expert statement of Richard James. Docket No.: 2525-CE-100. http://apps.psc.wi.gov/vs2015/ERF_ view/viewdoc.aspx?docid=284895. Accessed 14 Nov 2017

Journal

Acoustics AustraliaSpringer Journals

Published: Feb 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off